CN106686264A - Method and system for fraud call screening and analyzing - Google Patents
Method and system for fraud call screening and analyzing Download PDFInfo
- Publication number
- CN106686264A CN106686264A CN201610965273.9A CN201610965273A CN106686264A CN 106686264 A CN106686264 A CN 106686264A CN 201610965273 A CN201610965273 A CN 201610965273A CN 106686264 A CN106686264 A CN 106686264A
- Authority
- CN
- China
- Prior art keywords
- parameter
- caller
- call
- data
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/436—Arrangements for screening incoming calls, i.e. evaluating the characteristics of a call before deciding whether to answer it
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention belongs to the technical field of harmful call monitoring, and particularly relates to a method and system for fraud call screening and analyzing. According to the system, historical data is analyzed with a fraud call analysis model, and all feature weight values of the model are determined; real-time data is analyzed and detected, and a detection result is compared with a set threshold value to provide the confidence degree of a fraud call. The overall system is composed of a data query and management system, a real-time detection system, a model self-learning system, a trend prediction system and a data storage system. The data query and management system provides a full-amount call list query function, a fraud call list query function, a model parameter management function, a self-learning management function and a trend prediction analyzing function; the real-time detection system analyzes and detects call list data in real time through a fraud call discovery model to discover the fraud call; the model self-learning system analyzes historical call list data and continuously optimizes model parameters through a self-learning algorithm; the trend prediction system is used for predicting the future fraud call trend and change; the data storage system provides a quick data capturing function, a data distribution function and a data query function for the overall system by adopting a distributed storage system and a big data analysis processing engine.
Description
Technical field
The invention belongs to be harmful to phone-monitoring technical field in telecommunications, more particularly, to a kind of fraudulent call analysis side is screened
Method and system.
Background technology
At present telecommunications network fraudulent call finds to rely primarily on to exemplary call pattern analysis, such as " rings a sound ", " exhale dead you ",
" counterfeit number " etc., for specific call pattern comparison of classification analysis is carried out.This analytical technology can only be exhaled for specific typical case
Pattern is, and the harmful KOC kind of call of current telecommunications net emerges in an endless stream, means constantly update change, it is impossible to which realization automatically analyzes knowledge
Not.
The content of the invention
The characteristics of being continually changing for above-mentioned harmful calling means, cannot automatic identification in order to make up above-mentioned harmful phone
Defect, the present invention proposes a kind of fraudulent call screening assays and system.
A kind of fraudulent call screening assays, including preliminary screening and postsearch screening, by first screening and secondary sieve
Choosing obtains the doubtful number of fraudulent call;
The preliminary screening includes,
The Calling history of at least one analyzed number in time T is obtained from communication server;
Caller frequency parameter M1, the caller time interval parameter M2 of each analyzed number in calculating time t;Caller is conversed
Duration parameters M3, by counterfeit count parameter M4, calling number feature M5;
By caller frequency weight Q1, caller time interval weight Q2;Caller duration of call weight Q3, by counterfeit number of times weight
Q4, calling number feature weight Q5 and above-mentioned caller frequency parameter M1, caller time interval parameter M2;Caller duration of call parameter
M3, by counterfeit count parameter M4, calling number feature M5 combine, generate primary screener analysis model M=M1Q1+ of each phone
M2Q2+M3Q3+M4Q4+M5Q5;
Primary screener analysis model is compared with primary fiducial value K, when M > K, marks the phone can for primary
Doubtful telephone number, as M≤K, is then labeled as regular phone number;
The postsearch screening step includes,
The called history of the suspicious number in time T is obtained from communication server;
Extract called number characteristic parameter M6, the called network of personal connections characteristic parameter of the suspicious call number that primary screener goes out
M7, called type feature parameter M8;
By called number feature weight Q6, called number network of personal connections analysis feature weight Q7, called type feature weight Q8
Above-mentioned called number feature weight M6, called network of personal connections feature weight M7, called type feature weight M8 are substituted into, secondary sieve is obtained
Select analysis model m=M6Q6+M7Q7+M8Q8;
Secondary analysis model m is compared with secondary fiducial value k, as m > k, then fraudulent call is labeled as, as m≤k
When, then number is monitored for emphasis, the message registration of the number other times section is reacquired, repeat screening step above.
Further,
The computational methods of the caller frequency parameter M1 take advantage of frequency for analyzed number all caller number of times R1 in time T
Parameter T1, i.e. M1=R1T1;
The caller time interval parameter M2 computational methods are that analyzed number all caller interval times in time T are total
Spacing parameter T2, i.e. M2=R2T2 are taken advantage of with R2;
The caller duration of call parameter M3 computational methods are that analyzed number all caller air times in time T are total
Session parameter T3, i.e. M3=R3T3 are taken advantage of with R3;
The caller by counterfeit count parameter M4 computational methods be analyzed number in time T all callers by counterfeit time
Number R4 takes advantage of spacing parameter T4, i.e. M4=R4T4;
Further, the acquisition methods of the weight are:
It is assumed that the white list set W in data acquisition system E set E in blacklist set B, set E, respectively using each model pair
Set B and set W are calculated, and acquisition meets ratio of the number of model M X in list, wherein X={ 1,2,3,4,5 }.Symbol
The number of matched moulds type MX meets ratio WX of the number of model M X in set W, each Model Weight QX in ratio BX of set B
=BX-WX.
Further, the K=∑s BX- ∑ WX.
Further, the calling number feature M5 acquisition methods include that according to number length is classified to calling number,
It is assumed that number length set L is { L1, L2 ... LN }, Li ∈ L, the front Li-2 positions for taking number nm obtain nm ', by calling number nm
Difference, but nm ' identical numbers are classified as hundred segment number codes, and by that analogy, front Li-3 positions identical number is classified as kilosegment number, front
Li-4 positions identical number is classified as ten thousand segment number codes.
A kind of equipment using above-mentioned fraudulent call screening technique, the equipment includes:
Acquisition module, for obtaining the data of number talk times, time record from carrier server;
Storage module, the message registration of the number obtained for storage, the data of talk times;
Extraction module, for going out to cry frequency parameter M1, caller time interval parameter M2 from the extracting data of storage;Caller
Duration of call parameter M3, by counterfeit count parameter M4, calling number feature M5, called number feature weight Q6, called number close
System net analysis feature weight Q7 and called type feature weight Q8;
Primary calculations module, for calculating M=M1Q1+M2Q2+M3Q3+M4Q4+M5Q5;
Primary comparison module, for comparing with the value of K by M primary suspicious call number is drawn;
Secondary computing module, for calculating m=M6Q6+M7Q7+M8Q8;
Secondary comparison module, for comparing with the value of k by m suspicious call number is drawn.
The system of the present invention is analyzed using fraudulent call analysis model to historical data, determines each feature weight of model
Value;Detection is analyzed to real time data, testing result compares the confidence level for providing fraudulent call with given threshold.Whole system
It is made up of data query management system, real-time detecting system, Model Self-Learning system, trend predicting system, data-storage system.
Data query management system offer full dose CDR inquiry, swindle CDR inquiry, model parameter management, self study management, trend are pre-
Survey analytic function.Real-time detecting system is analyzed in real time by fraudulent call discovery model, detection ticket data, finds swindle electricity
Words.Model Self-Learning system is analyzed history call bill data, and by self-learning algorithm model parameter is continued to optimize.Trend prediction system
System provides and following fraudulent call trend and change is predicted.Data-storage system adopts distributed memory system, big data
Analyzing and processing engine provides rapid data crawl, data distribution, data query function for whole system.
Specific embodiment
Technical scheme will be clearly and completely described below, it is clear that described embodiment is this
Bright a part of embodiment, rather than the embodiment of whole.Based on the embodiment in the present invention, those of ordinary skill in the art are not having
Have and make the every other embodiment obtained under the premise of creative work, belong to the scope of protection of the invention.
In describing the invention, it should be noted that term " " center ", " on ", D score, "left", "right", " vertical ",
The orientation or position relationship of the instruction such as " level ", " interior ", " outward " be shown in orientation or position relationship, be for only for ease of description
The present invention is described with simplifying, rather than is indicated or implied that the device or element of indication must be with specific orientation, with specific
Azimuth configuration and operation, therefore be not considered as limiting the invention.Additionally, term " first ", " second ", " the 3rd " are only used
In description purpose, and it is not intended that indicating or implying relative importance.
In describing the invention, it should be noted that unless otherwise clearly defined and limited, term " installation ", " phase
Company ", " connection " should be interpreted broadly, for example, it may be being fixedly connected, or being detachably connected, or be integrally connected;Can
Being to be mechanically connected, or electrically connect;Can be joined directly together, it is also possible to be indirectly connected to by intermediary, Ke Yishi
The connection of two element internals.For the ordinary skill in the art, with concrete condition above-mentioned term can be understood at this
Concrete meaning in invention.
One embodiment of the present invention.
A kind of fraudulent call screening assays, including preliminary screening and postsearch screening, by first screening and secondary sieve
Choosing obtains the doubtful number of fraudulent call;
The preliminary screening includes,
The Calling history of at least one analyzed number in time T is obtained from communication server;
Caller frequency parameter M1, the caller time interval parameter M2 of each analyzed number in calculating time t;Caller is conversed
Duration parameters M3, by counterfeit count parameter M4, calling number feature M5;
By caller frequency weight Q1, caller time interval weight Q2;Caller duration of call weight Q3, by counterfeit number of times weight
Q4, calling number feature weight Q5 and above-mentioned caller frequency parameter M1, caller time interval parameter M2;Caller duration of call parameter
M3, by counterfeit count parameter M4, calling number feature M5 combine, generate primary screener analysis model M=M1Q1+ of each phone
M2Q2+M3Q3+M4Q4+M5Q5;
Primary screener analysis model is compared with primary fiducial value K, when M > K, marks the phone can for primary
Doubtful telephone number, as M≤K, is then labeled as regular phone number;
The postsearch screening step includes,
The called history of the suspicious number in time T is obtained from communication server;
Extract called number characteristic parameter M6, the called network of personal connections characteristic parameter of the suspicious call number that primary screener goes out
M7, called type feature parameter M8;
By called number feature weight Q6, called number network of personal connections analysis feature weight Q7, called type feature weight Q8
Above-mentioned called number feature weight M6, called network of personal connections feature weight M7, called type feature weight M8 are substituted into, secondary sieve is obtained
Select analysis model m=M6Q6+M7Q7+M8Q8;
Secondary analysis model m is compared with secondary fiducial value k, as m > k, then fraudulent call is labeled as, as m≤k
When, then number is monitored for emphasis, the message registration of the number other times section is reacquired, repeat screening step above.
Further,
The computational methods of the caller frequency parameter M1 take advantage of frequency for analyzed number all caller number of times R1 in time T
Parameter T1, i.e. M1=R1T1;
The caller time interval parameter M2 computational methods are that analyzed number all caller interval times in time T are total
Spacing parameter T2, i.e. M2=R2T2 are taken advantage of with R2;
The caller duration of call parameter M3 computational methods are that analyzed number all caller air times in time T are total
Session parameter T3, i.e. M3=R3T3 are taken advantage of with R3;
The caller by counterfeit count parameter M4 computational methods be analyzed number in time T all callers by counterfeit time
Number R4 takes advantage of spacing parameter T4, i.e. M4=R4T4;
Further, the acquisition methods of the weight are:
It is assumed that the white list set W in data acquisition system E, set E in blacklist set B, set E, respectively using each model
Set B and set W are calculated, acquisition meets ratio of the number of model M X in list, wherein X={ 1,2,3,4,5 }.
Meet ratio BX of the number in set B of model M X, meet ratio WX of the number of model M X in set W, each Model Weight
QX=BX-WX.
Further, the K=∑s BX-∑WX。
Further, the calling number feature M5 acquisition methods include that according to number length is classified to calling number,
It is assumed that number length set L is { L1, L2 ... LN }, Li ∈ L, the front Li-2 positions for taking number nm obtain nm ', by calling number nm
Difference, but nm ' identical numbers are classified as hundred segment number codes, and by that analogy, front Li-3 positions identical number is classified as kilosegment number, front
Li-4 positions identical number is classified as ten thousand segment number codes.
A kind of equipment using above-mentioned fraudulent call screening technique, the equipment includes:
Acquisition module, for obtaining the data of number talk times, time record from carrier server;
Storage module, the message registration of the number obtained for storage, the data of talk times;
Extraction module, for going out to cry frequency parameter M1, caller time interval parameter M2 from the extracting data of storage;Caller
Duration of call parameter M3, by counterfeit count parameter M4, calling number feature M5, called number feature weight Q6, called number close
System net analysis feature weight Q7 and called type feature weight Q8;
Primary calculations module, for calculating M=M1Q1+M2Q2+M3Q3+M4Q4+M5Q5;
Primary comparison module, for comparing with the value of K by M primary suspicious call number is drawn;
Secondary computing module, for calculating m=M6Q6+M7Q7+M8Q8;
Secondary comparison module, for comparing with the value of k by m suspicious call number is drawn.
The system of the present invention is analyzed using fraudulent call analysis model to historical data, determines each feature weight of model
Value;Detection is analyzed to real time data, testing result compares the confidence level for providing fraudulent call with given threshold.Whole system
It is made up of data query management system, real-time detecting system, Model Self-Learning system, trend predicting system, data-storage system.
Data query management system offer full dose CDR inquiry, swindle CDR inquiry, model parameter management, self study management, trend are pre-
Survey analytic function.Real-time detecting system is analyzed in real time by fraudulent call discovery model, detection ticket data, finds swindle electricity
Words.Model Self-Learning system is analyzed history call bill data, and by self-learning algorithm model parameter is continued to optimize.Trend prediction system
System provides and following fraudulent call trend and change is predicted.Data-storage system adopts distributed memory system, big data
Analyzing and processing engine provides rapid data crawl, data distribution, data query function for whole system.
Above-mentioned technical proposal only embodies the optimal technical scheme of technical solution of the present invention, those skilled in the art
Some of which part is made some variation embody the present invention principle, belong to protection scope of the present invention it
It is interior.
Claims (6)
1. a kind of fraudulent call screening assays, it is characterised in that including preliminary screening and postsearch screening, by first screening
With the doubtful number that postsearch screening obtains fraudulent call;
The preliminary screening includes,
The Calling history of at least one analyzed number in time T is obtained from communication server;
Caller frequency parameter M1, the caller time interval parameter M2 of each analyzed number in calculating time t;The caller duration of call
Parameter M3, by counterfeit count parameter M4, calling number feature M5;
By caller frequency weight Q1, caller time interval weight Q2;Caller duration of call weight Q3, by counterfeit number of times weight Q4,
Calling number feature weight Q5 and above-mentioned caller frequency parameter M1, caller time interval parameter M2;Caller duration of call parameter M3,
Combined by counterfeit count parameter M4, calling number feature M5, generate primary screener analysis model M=M1Q1+ of each phone
M2Q2+M3Q3+M4Q4+M5Q5;
Primary screener analysis model is compared with primary fiducial value K, when M > K, marks the phone to be primary suspicious electricity
Words number, as M≤K, is then labeled as regular phone number;
The postsearch screening step includes,
The called history of the suspicious number in time T is obtained from communication server;
Extract called number characteristic parameter M6, called network of personal connections characteristic parameter M7, the quilt of the suspicious call number that primary screener goes out
It is type feature parameter M8;
Called number feature weight Q6, called number network of personal connections analysis feature weight Q7, called type feature weight Q8 are substituted into
Above-mentioned called number feature weight M6, called network of personal connections feature weight M7, called type feature weight M8, obtain postsearch screening point
Analysis model m=M6Q6+M7Q7+M8Q8;
Secondary analysis model m is compared with secondary fiducial value k, as m > k, then fraudulent call is labeled as, as m≤k,
Then number is monitored for emphasis, reacquire the message registration of the number other times section, repeat screening step above.
2. fraudulent call screening assays according to claim 1, it is characterised in that
The computational methods of the caller frequency parameter M1 take advantage of frequency parameter for analyzed number all caller number of times R1 in time T
T1, i.e. M1=R1T1;
The caller time interval parameter M2 computational methods are analyzed number all caller interval time summations R2 in time T
Take advantage of spacing parameter T2, i.e. M2=R2T2;
The caller duration of call parameter M3 computational methods are analyzed number all caller air time summations R3 in time T
Take advantage of session parameter T3, i.e. M3=R3T3;
The caller by counterfeit count parameter M4 computational methods be analyzed number in time T all callers by counterfeit number of times R4
Take advantage of spacing parameter T4, i.e. M4=R4T4.
3. fraudulent call screening assays according to claim 2, it is characterised in that the acquisition methods of the weight
For:
It is assumed that the white list set W in data acquisition system E, set E in blacklist set B, set E, respectively using each model to collection
Close B and set W is calculated, acquisition meets ratio of the number of model M X in list, wherein X={ 1,2,3,4,5 }.Meet
The number of model M X meets ratio WX of the number of model M X in set W, each Model Weight QX=in ratio BX of set B
BX-WX。
4. fraudulent call screening assays according to claim 3, it is characterised in that the K=∑s BX- ∑ WX.
5. fraudulent call screening assays according to claim 4, it is characterised in that calling number feature M5 is obtained
The method of taking includes that according to number length is classified to calling number, it is assumed that number length set L be { L1, L2 ... LN }, Li ∈
L, the front Li-2 positions for taking number nm obtain nm ', and calling number nm is different, but nm ' identical numbers are classified as hundred segment number codes, with this
Analogize, front Li-3 positions identical number is classified as kilosegment number, front Li-4 positions identical number is classified as ten thousand segment number codes.
6. the system of fraudulent call screening technique described in a kind of employing claim 1, it is characterised in that the system includes:
Acquisition module, for obtaining the data of number talk times, time record from carrier server;
Storage module, the message registration of the number obtained for storage, the data of talk times;
Extraction module, for going out to cry frequency parameter M1, caller time interval parameter M2 from the extracting data of storage;Caller is conversed
Duration parameters M3, by counterfeit count parameter M4, calling number feature M5, called number feature weight Q6, called number network of personal connections
Analysis feature weight Q7 and called type feature weight Q8;
Primary calculations module, for calculating M=M1Q1+M2Q2+M3Q3+M4Q4+M5Q5;
Primary comparison module, for comparing with the value of K by M primary suspicious call number is drawn;
Secondary computing module, for calculating m=M6Q6+M7Q7+M8Q8;
Secondary comparison module, for comparing with the value of k by m suspicious call number is drawn.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610965273.9A CN106686264B (en) | 2016-11-04 | 2016-11-04 | Fraud telephone screening and analyzing method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610965273.9A CN106686264B (en) | 2016-11-04 | 2016-11-04 | Fraud telephone screening and analyzing method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106686264A true CN106686264A (en) | 2017-05-17 |
CN106686264B CN106686264B (en) | 2021-03-02 |
Family
ID=58839237
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610965273.9A Active CN106686264B (en) | 2016-11-04 | 2016-11-04 | Fraud telephone screening and analyzing method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106686264B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107342986A (en) * | 2017-06-20 | 2017-11-10 | 深圳安巽科技有限公司 | A kind of fraudulent call early warning system and method for early warning |
CN107770777A (en) * | 2017-09-30 | 2018-03-06 | 杭州东信北邮信息技术有限公司 | A kind of recognition methods for fraudulent call of recording |
CN108156336A (en) * | 2018-02-01 | 2018-06-12 | 天津市国瑞数码安全系统股份有限公司 | A kind of control system based on multiple features weighted model analysis ticket number |
CN108156333A (en) * | 2018-02-01 | 2018-06-12 | 天津市国瑞数码安全系统股份有限公司 | A kind of control system for preventing swindle |
CN108924333A (en) * | 2018-06-12 | 2018-11-30 | 阿里巴巴集团控股有限公司 | Fraudulent call recognition methods, device and system |
CN109379327A (en) * | 2018-04-13 | 2019-02-22 | 国家计算机网络与信息安全管理中心 | Based on manifold signaling list criticality analysis method and analysis system |
CN109493882A (en) * | 2018-11-04 | 2019-03-19 | 国家计算机网络与信息安全管理中心 | A kind of fraudulent call voice automatic marking system and method |
CN109510903A (en) * | 2018-11-16 | 2019-03-22 | 杭州东信北邮信息技术有限公司 | A method of the international swindle number of identification |
CN109756901A (en) * | 2017-11-06 | 2019-05-14 | 中国电信股份有限公司 | Anti- swindle method and apparatus |
CN110059889A (en) * | 2019-03-28 | 2019-07-26 | 国家计算机网络与信息安全管理中心 | Swindle calling sequence detection method based on unsupervised learning |
CN110248322A (en) * | 2019-06-28 | 2019-09-17 | 国家计算机网络与信息安全管理中心 | A kind of swindling gang identifying system and recognition methods based on fraud text message |
CN111915468A (en) * | 2020-08-14 | 2020-11-10 | 中国科学院自动化研究所 | Novel anti-fraud active inspection and early warning system for network |
CN113163057A (en) * | 2021-01-20 | 2021-07-23 | 北京工业大学 | Method for constructing dynamic identification interval of fraud telephone |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5970405A (en) * | 1997-02-28 | 1999-10-19 | Cellular Technical Services Co., Inc. | Apparatus and method for preventing fraudulent calls in a wireless telephone system using destination and fingerprint analysis |
CN101217820A (en) * | 2008-01-15 | 2008-07-09 | 中兴通讯股份有限公司 | An identification system and identification method on disturbance telephone numbers |
CN101257671A (en) * | 2007-07-06 | 2008-09-03 | 浙江大学 | Method for real time filtering large scale rubbish SMS based on content |
CN102368853A (en) * | 2011-09-26 | 2012-03-07 | 广州市动景计算机科技有限公司 | Communication event processing method and system thereof |
CN104038648A (en) * | 2014-06-06 | 2014-09-10 | 陈伟 | Method and device for recognizing crank calls |
CN104243727A (en) * | 2014-10-15 | 2014-12-24 | 上海欣方智能系统有限公司 | System and method for performing big data analysis confirmation and interception on phone scams |
CN104244216A (en) * | 2014-09-29 | 2014-12-24 | 中国移动通信集团浙江有限公司 | Method and system for intercepting fraud phones in real time during calling |
CN104469025A (en) * | 2014-11-26 | 2015-03-25 | 杭州东信北邮信息技术有限公司 | Clustering-algorithm-based method and system for intercepting fraud phone in real time |
CN104735671A (en) * | 2015-02-27 | 2015-06-24 | 腾讯科技(深圳)有限公司 | Malicious call recognition method and device |
CN105611084A (en) * | 2016-01-29 | 2016-05-25 | 中国联合网络通信集团有限公司 | User fraud suspiciousness degree calculation method and suspiciousness degree calculation system |
-
2016
- 2016-11-04 CN CN201610965273.9A patent/CN106686264B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5970405A (en) * | 1997-02-28 | 1999-10-19 | Cellular Technical Services Co., Inc. | Apparatus and method for preventing fraudulent calls in a wireless telephone system using destination and fingerprint analysis |
CN101257671A (en) * | 2007-07-06 | 2008-09-03 | 浙江大学 | Method for real time filtering large scale rubbish SMS based on content |
CN101217820A (en) * | 2008-01-15 | 2008-07-09 | 中兴通讯股份有限公司 | An identification system and identification method on disturbance telephone numbers |
CN102368853A (en) * | 2011-09-26 | 2012-03-07 | 广州市动景计算机科技有限公司 | Communication event processing method and system thereof |
CN104038648A (en) * | 2014-06-06 | 2014-09-10 | 陈伟 | Method and device for recognizing crank calls |
CN104244216A (en) * | 2014-09-29 | 2014-12-24 | 中国移动通信集团浙江有限公司 | Method and system for intercepting fraud phones in real time during calling |
CN104243727A (en) * | 2014-10-15 | 2014-12-24 | 上海欣方智能系统有限公司 | System and method for performing big data analysis confirmation and interception on phone scams |
CN104469025A (en) * | 2014-11-26 | 2015-03-25 | 杭州东信北邮信息技术有限公司 | Clustering-algorithm-based method and system for intercepting fraud phone in real time |
CN104735671A (en) * | 2015-02-27 | 2015-06-24 | 腾讯科技(深圳)有限公司 | Malicious call recognition method and device |
CN105611084A (en) * | 2016-01-29 | 2016-05-25 | 中国联合网络通信集团有限公司 | User fraud suspiciousness degree calculation method and suspiciousness degree calculation system |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107342986A (en) * | 2017-06-20 | 2017-11-10 | 深圳安巽科技有限公司 | A kind of fraudulent call early warning system and method for early warning |
CN107770777A (en) * | 2017-09-30 | 2018-03-06 | 杭州东信北邮信息技术有限公司 | A kind of recognition methods for fraudulent call of recording |
CN109756901A (en) * | 2017-11-06 | 2019-05-14 | 中国电信股份有限公司 | Anti- swindle method and apparatus |
CN108156336A (en) * | 2018-02-01 | 2018-06-12 | 天津市国瑞数码安全系统股份有限公司 | A kind of control system based on multiple features weighted model analysis ticket number |
CN108156333A (en) * | 2018-02-01 | 2018-06-12 | 天津市国瑞数码安全系统股份有限公司 | A kind of control system for preventing swindle |
CN109379327A (en) * | 2018-04-13 | 2019-02-22 | 国家计算机网络与信息安全管理中心 | Based on manifold signaling list criticality analysis method and analysis system |
CN108924333A (en) * | 2018-06-12 | 2018-11-30 | 阿里巴巴集团控股有限公司 | Fraudulent call recognition methods, device and system |
CN109493882A (en) * | 2018-11-04 | 2019-03-19 | 国家计算机网络与信息安全管理中心 | A kind of fraudulent call voice automatic marking system and method |
CN109510903A (en) * | 2018-11-16 | 2019-03-22 | 杭州东信北邮信息技术有限公司 | A method of the international swindle number of identification |
CN109510903B (en) * | 2018-11-16 | 2020-08-11 | 杭州东信北邮信息技术有限公司 | Method for identifying international fraud number |
CN110059889A (en) * | 2019-03-28 | 2019-07-26 | 国家计算机网络与信息安全管理中心 | Swindle calling sequence detection method based on unsupervised learning |
CN110059889B (en) * | 2019-03-28 | 2021-05-28 | 国家计算机网络与信息安全管理中心 | Fraud call sequence detection method based on unsupervised learning |
CN110248322A (en) * | 2019-06-28 | 2019-09-17 | 国家计算机网络与信息安全管理中心 | A kind of swindling gang identifying system and recognition methods based on fraud text message |
CN110248322B (en) * | 2019-06-28 | 2021-10-22 | 国家计算机网络与信息安全管理中心 | Fraud group partner identification system and identification method based on fraud short messages |
CN111915468A (en) * | 2020-08-14 | 2020-11-10 | 中国科学院自动化研究所 | Novel anti-fraud active inspection and early warning system for network |
CN111915468B (en) * | 2020-08-14 | 2021-09-10 | 中国科学院自动化研究所 | Network anti-fraud active inspection and early warning system |
CN113163057A (en) * | 2021-01-20 | 2021-07-23 | 北京工业大学 | Method for constructing dynamic identification interval of fraud telephone |
CN113163057B (en) * | 2021-01-20 | 2022-09-30 | 北京工业大学 | Method for constructing dynamic identification interval of fraud telephone |
Also Published As
Publication number | Publication date |
---|---|
CN106686264B (en) | 2021-03-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106686264A (en) | Method and system for fraud call screening and analyzing | |
CN106899435B (en) | A kind of complex attack recognition methods towards wireless invasive detection system | |
He et al. | Software-defined-networking-enabled traffic anomaly detection and mitigation | |
Zhang et al. | Random-forests-based network intrusion detection systems | |
CN104935600A (en) | Mobile ad hoc network intrusion detection method and device based on deep learning | |
Zhang et al. | Clustering algorithm-based data fusion scheme for robust cooperative spectrum sensing | |
Yang et al. | Deep learning and zero-day traffic classification: Lessons learned from a commercial-grade dataset | |
CN108632269A (en) | Detecting method of distributed denial of service attacking based on C4.5 decision Tree algorithms | |
Efstathopoulos et al. | Operational data based intrusion detection system for smart grid | |
CN107038449B (en) | Method and device for identifying fraudulent user | |
CN102957579A (en) | Network anomaly traffic monitoring method and device | |
Wang et al. | Def-ids: An ensemble defense mechanism against adversarial attacks for deep learning-based network intrusion detection | |
CN117113262B (en) | Network traffic identification method and system | |
Stiawan et al. | Ping flood attack pattern recognition using a K-means algorithm in an Internet of Things (IoT) network | |
Li et al. | Anomaly detection for cellular networks using big data analytics | |
Ashraf et al. | Intrusion detection system for SDN-enabled IoT networks using machine learning techniques | |
CN114169438A (en) | Telecommunication network fraud identification method, device, equipment and storage medium | |
KR20180130295A (en) | Apparatus for predicting failure of communication network and method thereof | |
Madhu et al. | IoT Network Attack Severity Classification | |
Lee et al. | AI-based network security enhancement for 5G industrial internet of things environments | |
Nguyen et al. | Towards improving explainability, resilience and performance of cybersecurity analysis of 5G/IoT networks (work-in-progress paper) | |
CN116192530A (en) | Unknown threat self-adaptive detection method based on deceptive defense | |
CN110443187A (en) | The recording method of characteristic information and device | |
Ndong et al. | Signal processing-based anomaly detection techniques: a comparative analysis | |
Shen et al. | Prior knowledge based advanced persistent threats detection for IoT in a realistic benchmark |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |