CN109951609A - A kind of malicious call number processing method and device - Google Patents
A kind of malicious call number processing method and device Download PDFInfo
- Publication number
- CN109951609A CN109951609A CN201711387908.2A CN201711387908A CN109951609A CN 109951609 A CN109951609 A CN 109951609A CN 201711387908 A CN201711387908 A CN 201711387908A CN 109951609 A CN109951609 A CN 109951609A
- Authority
- CN
- China
- Prior art keywords
- group
- calling
- numbers
- similarity
- conversational nature
- 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
Abstract
The embodiment of the present invention provides a kind of malicious call number processing method and device.The described method includes: obtaining the conversational nature of each calling number during the sampling period in malicious call number library；Several group's numbers are identified from each calling number in the number section group according to the similarity between each calling number in the number section group for each number section group；Wherein, each calling number top N number section having the same in number section group；Similarity in number section group between each calling number is determined according to the conversational nature of each calling number in number section group；The each group's number identified meets following condition: similarity of the group's number at least between another group's number in the number section group is higher than given threshold；The group's number identified out of each number section group is added to preset Call Intercept blacklist.Described device is for executing the above method.Method and apparatus provided in an embodiment of the present invention improve the accuracy of malicious call number interception.
Description
Technical field
The present embodiments relate to field of communication technology more particularly to a kind of malicious call number processing methods and device.
Background technique
Currently, occur on network a kind of software " exhale dead you " also known as the networking telephone chase after automatically paging system or " mobile phone Hong
Fried software ".The software is using the cheap networking telephone of communication fee as call platform, using international advanced networks telephonic communication
Technology can be convenient setting and chase after the fixed-line telephone and phone number for exhaling any one, any region.Criminal is by " exhaling dead
You " software initiates malicious call constantly to harass user or even blackmail.
It is existing to be identified and be intercepted by the way of user's mark for these malicious calls.User is disliked
After meaning is made nuisance calls, the calling number that can feed back its malicious call is malicious call number, Cloud Server or customer service to the master
Code of calling out the numbers carries out malicious call number mark.In this way, when the calling number initiates calling again, call processing server or use
Family terminal can first carry out blacklist matching, and whether the calling number for judging to initiate the calling is malicious call number, if the caller
Number has been labeled as malicious call number and then intercepts its calling.
However, this method relies primarily on customer complaint, and there are certain subjectivities for customer complaint, if directly being thrown according to user
It tells and is intercepted, be easy to appear and accidentally block.Therefore, it is necessary to which the malicious call number to customer complaint is further processed, to mention
The accuracy rate that high malicious call number intercepts.
Summary of the invention
For the defects in the prior art, the embodiment of the present invention provides a kind of malicious call number processing method and device,
Improve the accuracy of malicious call number interception.
On the one hand, the embodiment of the present invention provides a kind of malicious call number processing method, comprising:
Obtain the conversational nature of each calling number during the sampling period in malicious call number library；
For each number section group, according to the similarity between each calling number in the number section group, out of this number section group
Each calling number in identify several group's numbers；Wherein, each calling number preceding N having the same in the number section group
Position number section, N are the integer that value is greater than 1；Similarity in the number section group between each calling number is according to the number section group
What the conversational nature of interior each calling number determined；The each group's number identified meets following condition: group's number
Similarity at least between another group's number in the number section group is higher than given threshold；
The group's number identified out of each number section group is added to preset Call Intercept blacklist.
On the other hand, the embodiment of the present invention provides a kind of malicious call number processing unit, comprising:
Conversational nature obtains module, for obtaining each calling number in malicious call number library during the sampling period
Conversational nature；
Group's number identification module, for being directed to each number section group, according between each calling number in the number section group
Similarity, several group's numbers are identified from each calling number in the number section group；Wherein, each in the number section group
Calling number top N number section having the same, N are the integer that value is greater than 1；Phase in the number section group between each calling number
It like degree is determined according to the conversational nature of each calling number in the number section group；The each group's number identified meets such as
Lower condition: similarity of the group's number at least between another group's number in the number section group is higher than given threshold；
Malicious call number processing module, for the group's number identified out of each number section group to be added to preset exhale
It cries and intercepts blacklist.
Another aspect, the embodiment of the present invention provide a kind of electronic equipment, including processor, memory and bus, in which:
The processor, the memory complete mutual communication by bus；
The processor can call the computer program in memory, the step of to execute the above method.
In another aspect, the embodiment of the present invention provides a kind of computer readable storage medium, it is stored thereon with computer program,
The step of above method is realized when the program is executed by processor.
Malicious call number processing method and device provided in an embodiment of the present invention, by obtaining in malicious call number library
Each calling number conversational nature during the sampling period；For each number section group, according to each caller in the number section group
Similarity between number identifies several group's numbers from each calling number in the number section group；Each of identify
Group's number meets following condition: similarity of the group's number at least between another group's number in the number section group
Higher than given threshold；The group's number identified out of each number section group is added to preset Call Intercept blacklist, realization pair
The malicious call number of customer complaint is further processed, and improves the accuracy rate of malicious call number interception.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 shows the exemplary process diagram of malicious call number processing method according to an embodiment of the invention；
Fig. 2 shows spectral clustering classification results schematic diagrames according to an embodiment of the invention；
Fig. 3 shows the structural schematic diagram of malicious call number processing unit according to an embodiment of the invention；
Fig. 4 shows the entity structure schematic diagram of electronic equipment according to an embodiment of the invention.
Specific embodiment
Clear, complete description is carried out to technical solution of the present invention below with reference to attached drawing, it is clear that described implementation
Example is only a part of the embodiments of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, this field is general
Logical technical staff obtained all other embodiment without making creative work belongs to the present invention and is protected
The range of shield.
The terms such as " module " used in this application, " device " be intended to include with computer-related entity, such as it is but unlimited
In hardware, firmware, combination thereof, software or software in execution.For example, module can be, and it is not limited to: processing
Process, processor, object, executable program, the thread of execution, program and/or the computer run on device.For example, it counts
Calculating the application program run in equipment and this calculating equipment can be module.One or more modules can be located in execution
In one process and/or thread, a module can also be located on a computer and/or be distributed in two or more calculating
Between machine.
The technical solution that the invention will now be described in detail with reference to the accompanying drawings.
With reference to Fig. 1, it illustrates the exemplary streams of malicious call number processing method according to an embodiment of the invention
Cheng Tu.
As shown in Figure 1, malicious call number processing method provided in an embodiment of the present invention, may include steps of:
S110: the conversational nature of each calling number during the sampling period in malicious call number library is obtained.
In the embodiment of the present invention, it is stored with complained calling number gathered in advance in malicious call number library and is sampling
Call bill data in period.
Wherein, forms data includes: calling number, called number, calling number ownership place, called number if calling number
Calling time started of ownership place, the calling date of the every calling and every calling and calling such as end of calling time is related believes
Breath.Sampling period is configured according to actual needs by those skilled in the art, for example, can be set to one day, one week, one
Month etc..It will illustrate the embodiment of the present invention for one day below.
It, can be according to the calling number for each calling number in malicious call number library in the embodiment of the present invention
Call bill data, statistics obtains calling number conversational nature during the sampling period.Wherein, calling number is during the sampling period
Conversational nature include calling number corresponding conversational nature node under characteristic parameter several different.
Wherein, characteristic parameter may include following at least one: call the date, day talk times, day be called number, whether
For local number, whether had calling behavior, the same day had call, every within how much each periods in non-normal hours section
The average call duration of calling.Certainly, in practical application, characteristic parameter can be increased according to actual needs.
Correspondingly, calling number corresponding conversational nature node under characteristic parameter can join for calling number in this feature
Practical value under several, or the numerical value that practical value obtains after pretreatment.
For example, unique corresponding numerical value can be configured for all dates；For characteristic parameter-call date, caller number
Code corresponding conversational nature node under this feature parameter is the unique corresponding numerical value of actual call conversation date institute.For spy
Whether sign parameter-is local number, calling number corresponding conversational nature node under this feature parameter may include: be local
Number, non-local number.In practical application, local number can be indicated with 1, use 0 indicates non-local number.Feature is joined
Number-day talk times can preset it is one or more divide threshold values, be convenient for classifying.For example, calling number is in feature
Corresponding conversational nature node may include: less than or equal to 15 times, greater than 15 times and be less than or wait under parameter-day talk times
In 50 times, greater than 50 times.
S120: it is directed to each number section group, according to the similarity between each calling number in the number section group, from the number section
Several group's numbers are identified in each calling number in group, the similarity in number section group between each calling number is basis number
What the conversational nature of each calling number in section group determined.
Wherein, each calling number top N number section having the same in the number section group, N are the integer that value is greater than 1.
Similarity in the number section group between each calling number is true according to the conversational nature of each calling number in the number section group
Fixed.
In the embodiment of the present invention, each group's number identified out of number section group meets following condition a: group number
Similarity of the code at least between another group's number in the number section group is higher than given threshold.
Specifically, it can be first compared, be identified according to the top N number section of each calling number in malicious call number library
The calling number of top N number section having the same out, and the multiple calling numbers for the top N number section having the same that will identify that
It is divided into the same grouping, forms the corresponding number section group of the number section.In practical application, the value of N is by those skilled in the art
It is rule of thumb configured, for example, can be set to 7.
Then, for each number section group of division, the group number in the number section group can be identified according to such as under type
Code: it according to the conversational nature of each calling number in number section group, filters out at least one identical conversational nature node
Several calling numbers are as candidate group number；For each candidate group's number, if the candidate group number at least with
Similarity between one other candidate group number is higher than given threshold, then identifies that the candidate group number is the number section group
Interior group's number.
Optionally, for each number section group of division, the group in the number section group can be identified according to such as under type
Number:
Two calling numbers that similarity is higher than given threshold are chosen out of this number section group；By two calling numbers of selection
It is added in the corresponding group's set of numbers of the number section group as group's number, and by two calling numbers of selection from the number section
It is rejected in group.
Then, each calling number in the number section group is traversed to carry out the identification of group's number and update group's number collection
It closes: if the similarity in the calling number and group's set of numbers currently chosen between any one group's number is higher than institute
Given threshold is stated, then is added to the calling number as group's number in group's set of numbers to update the group number
Code collection is closed, and the calling number currently chosen is rejected out of this number section group, chooses next calling number out of this number section group
Carry out the identification of group's number；If similar between the calling number currently chosen and group's numbers all in group's set of numbers
Degree is below or is equal to given threshold, then the identification that next calling number carries out group's number is chosen out of this number section group.If
The calling number currently chosen is the last one calling number in the number section group, then terminates group's number in the number section group
Identification.
For example, choosing two calling number A1 and calling number A2 that similarity is higher than given threshold out of this number section group；
It is added to the calling number A1 of selection and calling number A2 as group's number in the corresponding group's set of numbers of the number section group,
And calling number A1 and calling number A2 is rejected out of this number section group.Then, judge remaining calling number in number section group
Whether quantity is 0, if the quantity of remaining calling number is 0 in number section group, terminates the knowledge of group's number in the number section group
Not.If the quantity of remaining calling number is not 0 in number section group, next calling number is randomly selected out of number section group and is carried out
The identification of group's number.
For example, the next calling number randomly selected is calling number A3, it is random from group's set of numbers (A1, A2)
Choose group's number；According to calling number A3 and the respective conversational nature of group's number currently chosen, caller number is determined
Similarity between code A3 and group's number of selection；If the similarity between calling number A3 and the group's number currently chosen
More than given threshold, then can be added to calling number A3 as group's number in group's set of numbers, after obtaining update
Group's set of numbers (A1, A2, A3).Meanwhile can reject calling number A3 out of number section group, under being chosen out of number section group
One calling number carries out the identification of group's number.
It, can be from collection if the similarity between calling number A3 and the group's number currently chosen is less than given threshold
Next group's number is chosen in group's set of numbers, and carries out the judgement of similarity.If calling number A3 and this collection chosen
Similarity between group's number is higher than given threshold, then can be added to group's number for calling number A3 as group's number
In set, updated group's set of numbers (A1, A2, A3) is obtained.Meanwhile calling number A3 can be picked out of number section group
It removes, the identification that next calling number carries out group's number is chosen out of number section group.If calling number A3 and group's set of numbers
Similarity in similarity and calling number A3 and group's set of numbers between middle group's number A1 between group number A2
It is below or is equal to given threshold, then chooses the identification that next calling number carries out group's number out of number section group.
Calling number A3 is added in group's set of numbers as group's number and obtains updated group's set of numbers
After (A1, A2, A3), before the identification that next calling number carries out group's number is randomly selected out of number section group, judgement number
Whether the quantity that section organizes interior remaining calling number is 0, if the quantity of remaining calling number is 0 in number section group, terminates this
The identification of group's number in number section group.It is random out of number section group if the quantity of remaining calling number is not 0 in number section group
Choose the identification that next calling number carries out group's number.
For example, the next calling number randomly selected is calling number A4, if calling number A4 and group's set of numbers
Similarity in (A1, A2, A3) between any one group's number is higher than given threshold, then can be identified as calling number A4
Group's number is added in group's set of numbers, obtains updated group's set of numbers (A1, A2, A3, A4).
In the embodiment of the present invention, the similarity between any two calling number can be determined according to such as under type:
For each characteristic parameter, two calling numbers corresponding conversational nature section under this feature parameter is determined
Similarity between point；It is similar between conversational nature node corresponding under each characteristic parameter according to two calling numbers
The accumulated value of degree determines the similarity in the number section group between each calling number.
It is described to be directed to each characteristic parameter in the embodiment of the present invention, determine two calling numbers under this feature parameter
Similarity between corresponding conversational nature node, comprising:
It is corresponding with the second conversational nature node to enter chain neighbor node according to the first conversational nature node corresponding each first
Each second enters the similarity between chain neighbor node, determines the first conversational nature node described under this feature parameter and described the
Similarity between two call characteristic nodes.
Wherein, the first conversational nature node is that first calling number in two calling numbers is right under this feature parameter
The conversational nature node answered；Second conversational nature node is second calling number in two calling numbers in this feature parameter
Under corresponding conversational nature node；First enters chain neighbor node to have the first conversational nature section in malicious call number library
The calling number of point；Second enters chain neighbor node to have the caller of the second conversational nature node in malicious call number library
Number.
Similarity and two different caller numbers in the embodiment of the present invention, between two different conversational nature nodes
Similarity between code is dynamic change.In the initial procedure for determining similarity, two different conversational nature nodes it
Between similarity take the similarity between the similarity between the first initial value and two different calling numbers to take at the beginning of first
Initial value, and the first initial value is specially 0.Correspondingly, the similarity between same conversational nature node takes the second initial value, same
Similarity between calling number takes the second initial value, and the second initial value is specially 1.
In practical application, 1 two calling number A can be determined according to the following formulapAnd AqBetween similarity R (Ap,
Aq):
In formula 1, L (Ap) it is calling number ApConversational nature, including | L (Ap) | a conversational nature node, Li(Ap) refer to
Be L (Ap) in i-th of conversational nature node；L(Aq) it is calling number AqConversational nature, including | L (Aq) | a call is special
Levy node, Lj(Aq) refer to L (Aq) in j-th of conversational nature node；r(Li(Ap),Lj(Aq)) refer to conversational nature node Li
(Ap) and conversational nature node Lj(Aq) between similarity, i be value [1, | L (Ap) |] integer, j be value [1, | L (Aq)
|] integer.C is preset diffusion coefficient, is rule of thumb configured by those skilled in the art, for example, c=0.6 or c=
0.8。
In practical application, 2 two conversational nature node K can be determined according to the following formulaiAnd KiBetween similarity r
(Ki,Kj):
In formula 2, l (Ki) refer to i-th of conversational nature node K in the conversational nature of a calling numberiEnter chain neighbour
Occupy node set, l (Ki) in include malicious call number library in it is all have conversational nature node KiCalling number, each tool
There is conversational nature node KiCalling number as conversational nature node KiCorresponding one enters chain neighbor node；|l(Ki) | refer to
It is conversational nature node KiThe corresponding number for entering chain neighbor node；lp(Ki) refer to l (Ki) in p-th of calling number.l
(Kj) refer to j-th of conversational nature node K in the conversational nature of a calling numberjEnter chain neighbor node set, l (Kj)
In include malicious call number library in it is all have conversational nature node KjCalling number, each have conversational nature node Kj
Calling number as conversational nature node KjCorresponding one enters chain neighbor node；|l(Kj) | refer to conversational nature node Kj
It is corresponding enter chain part node number；lq(Kj) refer to l (Kj) in q-th of calling number.R(lp(Ki),lq(Kj)) refer to
It is calling number lp(Ki) and calling number lq(Kj) between similarity.C is preset diffusion coefficient, by those skilled in the art
Member is rule of thumb configured, for example, c=0.6 or c=0.8.
S130: the group's number identified out of each number section group is added to preset Call Intercept blacklist.
Wherein, the calling that calling number is initiated in the Call Intercept blacklist will be intercepted.
Malicious call number processing method provided in an embodiment of the present invention, by obtaining each of malicious call number library
The conversational nature of calling number during the sampling period；For each number section group, according to each calling number in the number section group it
Between similarity, several group's numbers are identified from each calling number in the number section group；The each group number identified
Code meets following condition: similarity of the group's number at least between another group's number in the number section group, which is higher than, to be set
Determine threshold value；The group's number identified out of each number section group is added to preset Call Intercept blacklist, realizes and user is thrown
The malicious call number told is further processed, and improves the accuracy rate of malicious call number interception.
On the basis of the above embodiments, in the malicious call number processing method that further embodiment of this invention provides, from
After identifying several group's numbers in each calling number in each number section group, further includes:
It is special according to the call of each non-group's number in malicious call number library using preset spectral clustering
Sign, clusters each non-group's number, obtains in k Lei Fei group number, the class center of all kinds of non-group's numbers and class
The corresponding conversational nature of the heart；K is the integer that value is greater than 1；
According to the conversational nature at the class center, the danger level of all kinds of non-group's numbers is identified；The danger level tool
Body be it is following any one: it is highly dangerous, poor risk, low degree of hazard, dangerous undetermined.
Specifically, identified from each calling number in each number section group by step S120 several group's numbers it
Afterwards, data point X={ s can be constructed1, s2... si... sn, siIt indicates i-th of non-group's number in malicious call number library, takes
Value is the integer of [1, n], and n is the quantity of non-group's number in malicious call number library；According in malicious call number library
The conversational nature of each non-group's number constructs similar matrix S.Wherein, each element in similar matrix S can according to the following formula 3
Definition:
In formula 3, siIndicate i-th of non-group's number in malicious call number library, sjIt indicates in malicious call number library
J-th of non-group's number, d (si, sj) indicate between i-th of non-group's number and the j-th non-group's number it is European away from
From σ is standard deviation.
Then, professional etiquette generalized, construction standardization similar matrix S' are carried out to similar matrix.For example, can be according to following public affairs
Formula 4 carries out professional etiquette model:
According to standardization similar matrix S', diagonal matrix D is constructed；And it is based on diagonal matrix D and standardization similar matrix S',
Construct Laplacian Matrix P.For example, can be with according to the following formula 5 construction Laplacian Matrixes:
P=D-1/2S'D-1/2(formula 5)
Then, the characteristic value of standardization similar matrix S' is calculated, and sequence arranges by size, is denoted as λ1≥λ2≥…≥λn,
Calculate characteristic gap sequence { g1, g2..., gn-1|gi=λi-λi-1, the maximum value of characteristic gap is sought, g is denoted ask, then of class
Number is k.
Seek feature vector corresponding to the k maximum eigenvalue of Laplacian Matrix P: v1, v2..., vk, structural matrix V=
[v1, v2..., vk)∈Rn×k, wherein vl(l=1,2 ..., k) is column vector, and n is the quantity of non-group's number.
The row vector of Standard Process V is denoted as matrix Y.For example, professional etiquette model can be carried out with according to the following formula 6:
In formula 6, vijFor the i-th row jth column element in matrix V.
Each row element of matrix Y is seen as space RkIn a point, by k mean algorithm by these point minute
Class；If the i-th row element of matrix Y belongs to jth class, corresponding calling number siBelong to jth class.
According to the conversational nature at all kinds of centers, the danger level of all kinds of non-group's numbers is identified.Wherein, danger level is specific
For it is following any one: it is highly dangerous, poor risk, low degree of hazard, dangerous undetermined.It, can be by this field skill in practical application
Art personnel carry out according to pre-set recognition strategy.For example, the day call time in the conversational nature at certain a kind of class center
Number be higher than other classes classes conversational natures in day talk times or calling period number be greater than setting threshold value, then
The danger level of such group's number can be identified for height.With reference to Fig. 2, it illustrates spectrum according to an embodiment of the invention is poly-
Class algorithm classification result schematic diagram.
Further, in the malicious call number processing method that further embodiment of this invention provides, all kinds of non-groups are identified
After the danger level of number, all non-group's numbers that danger level is highly dangerous can also be added to the calling and blocked
Blacklist is cut, and is that poor risk, low degree of hazard and dangerous all non-group's numbers undetermined are stored to default by danger level
Suspicious number library in.
Further, for the non-group's number in each of the suspicious number library, if this is non-in the specified observation period
Group's number is not complained, then identifies that non-group's number is right number；If non-group's number in the specified observation period
Danger level be identified as the danger level of highly dangerous or non-group's number and be identified as the number of poor risk and be more than
Non- group's number is then added to the Call Intercept blacklist by the frequency threshold value of setting.
Other steps of the embodiment of the present invention are similar to previous embodiment step, and the embodiment of the present invention repeats no more.
Malicious call number processing method provided in an embodiment of the present invention, by the non-group in malicious call number library
Number is classified, and different types of calling number in malicious call number library is added to Call Intercept blacklist or suspicious number
It in code library, number can be reduced accidentally blocks and be blocked with leakage, improve the accuracy rate that malicious call number intercepts.
On the basis of the various embodiments described above, further embodiment of this invention provides a kind of malicious call number processing dress
It sets.
With reference to Fig. 3, it illustrates the structural representations of malicious call number processing unit according to an embodiment of the invention
Figure.
As shown in figure 3, malicious call number processing unit 300 provided in an embodiment of the present invention may include: conversational nature
Obtain module 301, group's number identification module 302 and malicious call number processing module 303.
Wherein, each calling number that conversational nature obtains that module 301 is used to obtain in malicious call number library is sampling
Conversational nature in period.
Group's number identification module 302 be used for be directed to each number section group, according to each calling number in the number section group it
Between similarity, several group's numbers are identified from each calling number in the number section group.
Wherein, each calling number top N number section having the same in the number section group, N are the integer that value is greater than 1；
Similarity in the number section group between each calling number is true according to the conversational nature of each calling number in the number section group
Fixed；The each group's number identified meets following condition: group's number at least with another collection in the number section group
Similarity between group's number is higher than given threshold.
Malicious call number processing module 303 is preset for the group's number identified out of each number section group to be added to
Call Intercept blacklist.
Optionally, group's number identification module 302 is specifically used for choosing similarity out of this number section group higher than given threshold
Two calling numbers；The corresponding group's number collection of the number section group is added to using two calling numbers of selection as group's number
In conjunction, and two calling numbers of selection are rejected out of this number section group；Each calling number traversed in the number section group is collected
The identification of group's number, and update group's set of numbers: if in the calling number currently chosen and group's set of numbers
Similarity between any one group's number is higher than the given threshold, then is added to the calling number as group's number
To update group's set of numbers in group's set of numbers, and the calling number currently chosen is picked out of this number section group
It removes, the identification that next calling number carries out group's number is chosen out of this number section group；If the calling number currently chosen and institute
It states the similarity in group's set of numbers between all group's numbers and is below or is equal to given threshold, then selected out of this number section group
Next calling number is taken to carry out the identification of group's number.
In the embodiment of the present invention, the conversational nature of the calling number includes the calling number in spy several different
Levy corresponding conversational nature node under parameter.
Optionally, group's number identification module 302 be specifically used for according to such as under type determine any two calling number it
Between similarity: be directed to each characteristic parameter, determine two calling numbers corresponding call spy under this feature parameter
Levy the similarity between node；According to two calling numbers between conversational nature node corresponding under each characteristic parameter
The accumulated value of similarity determines the similarity in the number section group between each calling number.
Optionally, group's number identification module 302 is specifically used for being entered according to the first conversational nature node corresponding each first
Chain neighbor node corresponding with the second conversational nature node each second enters the similarity between chain neighbor node, determines in this feature
Similarity under parameter between the first conversational nature node and the second conversational nature node；
Wherein, the first conversational nature node is first calling number in two calling numbers in this feature parameter
Under corresponding conversational nature node；The second conversational nature node is second calling number in two calling numbers at this
Corresponding conversational nature node under characteristic parameter；Described first enters chain neighbor node to have institute in malicious call number library
State the calling number of the first conversational nature node；Described second enters chain neighbor node to have institute in malicious call number library
State the calling number of the second conversational nature node.
Optionally, malicious call number processing unit 300 can further include: the number sorted module of malicious call.
The number sorted module of malicious call is used to utilize preset spectral clustering, according in malicious call number library
Each non-group's number conversational nature, each non-group's number is clustered, k Lei Fei group number, all kinds of non-collection are obtained
The corresponding conversational nature in class center and class center of group's number；K is the integer that value is greater than 1；According to the logical of the class center
Feature is talked about, identifies the danger level of all kinds of non-group's numbers；The danger level be specially it is following any one: highly dangerous,
Poor risk, low degree of hazard, danger are undetermined.
Optionally, malicious call number processing module 303 is also used to all non-groups by danger level for highly dangerous
Number is added to the Call Intercept blacklist；It is poor risk, low degree of hazard and dangerous undetermined all non-by danger level
Group's number is stored into preset suspicious number library.
Optionally, malicious call number processing module 303 is also used to for each of the suspicious number library non-group
Number identifies that non-group's number is right number if non-group's number is not complained in the specified observation period；If referring to
The danger level of non-group's number is identified as the danger level of highly dangerous or non-group's number in the fixed observation period
The number for being identified as poor risk is more than the frequency threshold value of setting, then it is black non-group's number to be added to the Call Intercept
List.
Malicious call number processing unit provided in an embodiment of the present invention, by obtaining each of malicious call number library
The conversational nature of calling number during the sampling period；For each number section group, according to each calling number in the number section group it
Between similarity, several group's numbers are identified from each calling number in the number section group；The each group number identified
Code meets following condition: similarity of the group's number at least between another group's number in the number section group, which is higher than, to be set
Determine threshold value；The group's number identified out of each number section group is added to preset Call Intercept blacklist, realizes and user is thrown
The malicious call number told is further processed, and improves the accuracy rate of malicious call number interception.
It is real that the embodiment of malicious call number processing unit provided by the invention specifically can be used for executing above-mentioned each method
The process flow of example is applied, details are not described herein for function, is referred to the detailed description of above method embodiment.
With reference to Fig. 4, it illustrates the entity structure schematic diagrames of electronic equipment according to an embodiment of the invention.Such as Fig. 4
Shown, which may include: processor (processor) 401, memory (memory) 402 and bus 403,
In, processor 401, memory 402 completes mutual communication by bus 403.Processor 401 can call memory 402
In computer program, to execute method provided by above-mentioned each method embodiment, for example,
Obtain the conversational nature of each calling number during the sampling period in malicious call number library；
For each number section group, according to the similarity between each calling number in the number section group, out of this number section group
Each calling number in identify several group's numbers；Wherein, each calling number preceding N having the same in the number section group
Position number section, N are the integer that value is greater than 1；Similarity in the number section group between each calling number is according to the number section group
What the conversational nature of interior each calling number determined；The each group's number identified meets following condition: group's number
Similarity at least between another group's number in the number section group is higher than given threshold；
The group's number identified out of each number section group is added to preset Call Intercept blacklist.
In another embodiment, following method is realized when the processor 401 executes the computer program:
It is described to be directed to each number section group, according to the similarity between each calling number in the number section group, from the number section
Several group's numbers are identified in each calling number in group, comprising:
Two calling numbers that similarity is higher than given threshold are chosen out of this number section group；
It is added to two calling numbers of selection as group's number in the corresponding group's set of numbers of the number section group, and
Two calling numbers of selection are rejected out of this number section group；
Each calling number traversed in the number section group carries out the identification of group's number, and updates group's set of numbers:
If the similarity in the calling number and group's set of numbers currently chosen between any one group's number is higher than described
Given threshold is then added to the calling number as group's number in group's set of numbers to update group's number
Set, and the calling number currently chosen is rejected out of this number section group, chosen out of this number section group next calling number into
The identification of row group number；If the phase between the calling number currently chosen and all group's numbers in group's set of numbers
Given threshold is below or be equal to like degree, then the identification that next calling number carries out group's number is chosen out of this number section group.
In another embodiment, following method is realized when the processor 401 executes the computer program:
The conversational nature of the calling number includes that the calling number is right respectively under characteristic parameter several different
The conversational nature node answered；And
Similarity between any two calling number is determined according to such as under type:
For each characteristic parameter, two calling numbers corresponding conversational nature section under this feature parameter is determined
Similarity between point；
According to the tired of similarity of two calling numbers between conversational nature node corresponding under each characteristic parameter
It is value added, determine the similarity in the number section group between each calling number.
In another embodiment, following method is realized when the processor 401 executes the computer program:
It is described to be directed to each characteristic parameter, determine that the corresponding call under this feature parameter of two calling numbers is special
Levy the similarity between node, comprising:
It is corresponding with the second conversational nature node to enter chain neighbor node according to the first conversational nature node corresponding each first
Each second enters the similarity between chain neighbor node, determines the first conversational nature node described under this feature parameter and described the
Similarity between two call characteristic nodes；
Wherein, the first conversational nature node is first calling number in two calling numbers in this feature parameter
Under corresponding conversational nature node；The second conversational nature node is second calling number in two calling numbers at this
Corresponding conversational nature node under characteristic parameter；Described first enters chain neighbor node to have institute in malicious call number library
State the calling number of the first conversational nature node；Described second enters chain neighbor node to have institute in malicious call number library
State the calling number of the second conversational nature node.
In another embodiment, following method is realized when the processor 401 executes the computer program:
It is special according to the call of each non-group's number in malicious call number library using preset spectral clustering
Sign, clusters each non-group's number, obtains in k Lei Fei group number, the class center of all kinds of non-group's numbers and class
The corresponding conversational nature of the heart；K is the integer that value is greater than 1；
According to the conversational nature at the class center, the danger level of all kinds of non-group's numbers is identified；The danger level tool
Body be it is following any one: it is highly dangerous, poor risk, low degree of hazard, dangerous undetermined.
In another embodiment, following method is realized when the processor 401 executes the computer program:
All non-group's numbers that danger level is highly dangerous are added to the Call Intercept blacklist；
By danger level be poor risk, low degree of hazard and dangerous all non-group's numbers undetermined store to it is preset can
It doubts in number library.
In another embodiment, following method is realized when the processor 401 executes the computer program:
For the non-group's number in each of the suspicious number library, if non-group's number is not in the specified observation period
It is complained, then identify that non-group's number is right number；
If the danger level of non-group's number is identified as highly dangerous or the non-group number in the specified observation period
The number that the danger level of code is identified as poor risk is more than the frequency threshold value set, then non-group's number is added to institute
State Call Intercept blacklist.
Electronic equipment provided in an embodiment of the present invention at least has following technical effect that by obtaining malicious call number
The conversational nature of each calling number during the sampling period in library；For each number section group, according to each in the number section group
Similarity between calling number identifies several group's numbers from each calling number in the number section group；One group
Similarity of the number at least between another group's number in the number section group is higher than given threshold；It will know out of each number section group
Not Chu group's number be added to preset Call Intercept blacklist, realize to the further of the malicious call number of customer complaint
Processing improves the accuracy rate of malicious call number interception.
The embodiment of the present invention discloses a kind of computer program product, and the computer program product is non-transient including being stored in
Computer program on computer readable storage medium, the computer program include program instruction, when described program instructs quilt
When computer executes, computer is able to carry out method provided by above-mentioned each method embodiment, for example,
Obtain the conversational nature of each calling number during the sampling period in malicious call number library；For each number
Section group is identified from each calling number in the number section group according to the similarity between each calling number in the number section group
Several group's numbers；Wherein, each calling number top N number section having the same in the number section group, N are that value is greater than 1
Integer；Similarity in the number section group between each calling number is the call according to each calling number in the number section group
What feature determined；Similarity of one group's number at least between another group's number in the number section group is higher than setting threshold
Value；The group's number identified out of each number section group is added to preset Call Intercept blacklist.
The embodiment of the present invention provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage
Medium storing computer program, the computer program make the computer execute side provided by above-mentioned each method embodiment
Method, for example,
Obtain the conversational nature of each calling number during the sampling period in malicious call number library；For each number
Section group is identified from each calling number in the number section group according to the similarity between each calling number in the number section group
Several group's numbers；Wherein, each calling number top N number section having the same in the number section group, N are that value is greater than 1
Integer；Similarity in the number section group between each calling number is the call according to each calling number in the number section group
What feature determined；Similarity of one group's number at least between another group's number in the number section group is higher than setting threshold
Value；The group's number identified out of each number section group is added to preset Call Intercept blacklist.
In addition, the logical order in above-mentioned memory can be realized and as independence by way of SFU software functional unit
Product when selling or using, can store in a computer readable storage medium.Based on this understanding, of the invention
Technical solution substantially the part of the part that contributes to existing technology or the technical solution can be with software in other words
The form of product embodies, which is stored in a storage medium, including some instructions use so that
One computer installation (can be personal computer, server or network equipment etc.) executes each embodiment institute of the present invention
State all or part of the steps of method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can be with
Store the medium of program code.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer installation (can be personal computer, server or network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of malicious call number processing method characterized by comprising
Obtain the conversational nature of each calling number during the sampling period in malicious call number library；
For each number section group, according to the similarity between each calling number in the number section group, from each in the number section group
Several group's numbers are identified in calling number；Wherein, each calling number top N number having the same in the number section group
Section, N are the integer that value is greater than 1；Similarity in the number section group between each calling number is according in the number section group
What the conversational nature of each calling number determined；The each group's number identified meets following condition: group's number is at least
Similarity between another group's number in the number section group is higher than given threshold；
The group's number identified out of each number section group is added to preset Call Intercept blacklist.
2. the method according to claim 1, wherein described be directed to each number section group, according in the number section group
Each calling number between similarity, several group's numbers are identified from each calling number in the number section group, comprising:
Two calling numbers that similarity is higher than given threshold are chosen out of this number section group；
It is added to two calling numbers of selection as group's number in the corresponding group's set of numbers of the number section group, and will choosing
Two calling numbers taken are rejected out of this number section group；
Each calling number traversed in the number section group carries out the identification of group's number, and updates group's set of numbers, specifically
If including: that similarity in the calling number currently chosen and group's set of numbers between any one group's number is higher than
The given threshold is then added to the calling number as group's number in group's set of numbers to update the group
Set of numbers, and the calling number currently chosen is rejected out of this number section group, next caller number is chosen out of this number section group
Code carries out the identification of group's number；If in the calling number and group's set of numbers currently chosen between all group's numbers
Similarity be below or be equal to given threshold, then the knowledge that next calling number carries out group's number is chosen out of this number section group
Not.
3. method according to claim 1 or 2, which is characterized in that the conversational nature of the calling number includes the master
It calls out the numbers code corresponding conversational nature node under characteristic parameter several different；And
Similarity between any two calling number is determined according to such as under type:
For each characteristic parameter, determine two calling numbers under this feature parameter corresponding conversational nature node it
Between similarity；
According to the accumulated value of similarity of two calling numbers between conversational nature node corresponding under each characteristic parameter,
Determine the similarity in the number section group between each calling number.
4. according to the method described in claim 3, it is characterized in that, it is described be directed to each characteristic parameter, determine two callers
Similarity of the number between conversational nature node corresponding under this feature parameter, comprising:
Enter chain neighbor node corresponding with the second conversational nature node each according to the first conversational nature node corresponding each first
Two enter the similarity between chain neighbor node, determine that the first conversational nature node described under this feature parameter is logical with described second
Talk about the similarity between characteristic node；
Wherein, the first conversational nature node is that first calling number in two calling numbers is right under this feature parameter
The conversational nature node answered；The second conversational nature node is second calling number in two calling numbers in this feature
Corresponding conversational nature node under parameter；Described first to enter chain neighbor node be to have described the in malicious call number library
The calling number of one conversational nature node；Described second to enter chain neighbor node be to have described the in malicious call number library
The calling number of two call characteristic nodes.
5. method according to claim 1 to 4, which is characterized in that the method also includes:
It is right according to the conversational nature of each non-group's number in malicious call number library using preset spectral clustering
Each non-group's number is clustered, and the class center and class center pair of k Lei Fei group number, all kinds of non-group's numbers are obtained
The conversational nature answered；K is the integer that value is greater than 1；
According to the conversational nature at the class center, the danger level of all kinds of non-group's numbers is identified；The danger level is specially
It is following any one: it is highly dangerous, poor risk, low degree of hazard, dangerous undetermined.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
All non-group's numbers that danger level is highly dangerous are added to the Call Intercept blacklist；
It is that poor risk, low degree of hazard and dangerous all non-group's numbers undetermined are stored to preset suspicious number by danger level
In code library.
7. according to the method described in claim 6, it is characterized in that, the method also includes:
For the non-group's number in each of the suspicious number library, if non-group's number is not thrown in the specified observation period
It tells, then identifies that non-group's number is right number；
If the danger level of interior non-group's number of specified observation period is identified as highly dangerous or non-group's number
The number that danger level is identified as poor risk is more than the frequency threshold value set, then non-group's number is added to described exhale
It cries and intercepts blacklist.
8. a kind of malicious call number processing unit characterized by comprising
Conversational nature obtains module, for obtaining the call of each calling number in malicious call number library during the sampling period
Feature；
Group's number identification module, for being directed to each number section group, according to the phase between each calling number in the number section group
Like degree, several group's numbers are identified from each calling number in the number section group；Wherein, each caller in the number section group
Number top N number section having the same, N are the integer that value is greater than 1；Similarity in the number section group between each calling number
It is to be determined according to the conversational nature of each calling number in the number section group；The each group's number identified meets following item
Part: similarity of the group's number at least between another group's number in the number section group is higher than given threshold；
Malicious call number processing module is blocked for the group's number identified out of each number section group to be added to preset calling
Cut blacklist.
9. a kind of electronic equipment, which is characterized in that including processor, memory and bus, in which:
The processor, the memory complete mutual communication by bus；
The processor can call the computer program in memory, to execute as described in claim 1-7 any one
The step of method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
It realizes when execution such as the step of claim 1-7 any one the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711387908.2A CN109951609B (en) | 2017-12-20 | 2017-12-20 | Malicious telephone number processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711387908.2A CN109951609B (en) | 2017-12-20 | 2017-12-20 | Malicious telephone number processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109951609A true CN109951609A (en) | 2019-06-28 |
| CN109951609B CN109951609B (en) | 2021-07-23 |
Family
ID=67005355
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711387908.2A Active CN109951609B (en) | 2017-12-20 | 2017-12-20 | Malicious telephone number processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109951609B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110602304A (en) * | 2019-09-17 | 2019-12-20 | 卓尔智联(武汉)研究院有限公司 | Information processing method, device and storage medium |
| CN113364764A (en) * | 2021-06-02 | 2021-09-07 | 中国移动通信集团广东有限公司 | Information security protection method and device based on big data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080141332A1 (en) * | 2006-12-11 | 2008-06-12 | International Business Machines Corporation | System, method and program product for identifying network-attack profiles and blocking network intrusions |
| CN101472007A (en) * | 2007-12-28 | 2009-07-01 | 中国移动通信集团公司 | Method and system for determining disturbance telephone |
| US20160021246A1 (en) * | 2006-11-14 | 2016-01-21 | Microsoft Technology Licensing, Llc | Secured communication via location awareness |
| CN105704719A (en) * | 2014-11-28 | 2016-06-22 | 中国移动通信集团公司 | Method and apparatus for realizing monitoring strategy optimization of nuisance calls |
| CN106954218A (en) * | 2017-03-15 | 2017-07-14 | 中国联合网络通信集团有限公司 | The number sorted methods, devices and systems of one kind harassing and wrecking |
| CN107404589A (en) * | 2017-08-10 | 2017-11-28 | 北京泰迪熊移动科技有限公司 | Kind identification method, device and the terminal device of call number |
-
2017
- 2017-12-20 CN CN201711387908.2A patent/CN109951609B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160021246A1 (en) * | 2006-11-14 | 2016-01-21 | Microsoft Technology Licensing, Llc | Secured communication via location awareness |
| US20080141332A1 (en) * | 2006-12-11 | 2008-06-12 | International Business Machines Corporation | System, method and program product for identifying network-attack profiles and blocking network intrusions |
| CN101472007A (en) * | 2007-12-28 | 2009-07-01 | 中国移动通信集团公司 | Method and system for determining disturbance telephone |
| CN105704719A (en) * | 2014-11-28 | 2016-06-22 | 中国移动通信集团公司 | Method and apparatus for realizing monitoring strategy optimization of nuisance calls |
| CN106954218A (en) * | 2017-03-15 | 2017-07-14 | 中国联合网络通信集团有限公司 | The number sorted methods, devices and systems of one kind harassing and wrecking |
| CN107404589A (en) * | 2017-08-10 | 2017-11-28 | 北京泰迪熊移动科技有限公司 | Kind identification method, device and the terminal device of call number |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110602304A (en) * | 2019-09-17 | 2019-12-20 | 卓尔智联(武汉)研究院有限公司 | Information processing method, device and storage medium |
| CN110602304B (en) * | 2019-09-17 | 2021-06-11 | 卓尔智联(武汉)研究院有限公司 | Information processing method, device and storage medium |
| CN113364764A (en) * | 2021-06-02 | 2021-09-07 | 中国移动通信集团广东有限公司 | Information security protection method and device based on big data |
| CN113364764B (en) * | 2021-06-02 | 2022-07-12 | 中国移动通信集团广东有限公司 | Information security protection method and device based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109951609B (en) | 2021-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110401779A (en) | A kind of method, apparatus and computer readable storage medium identifying telephone number | |
| CN106686261A (en) | Information processing method and system | |
| CN110363076A (en) | Personal information correlating method, device and terminal device | |
| CN103971134B (en) | Image classification, retrieval and bearing calibration, and related device | |
| CN105446741B (en) | A kind of mobile applications discrimination method compared based on API | |
| CN107306306A (en) | Communicating number processing method and processing device | |
| CN109688275A (en) | Harassing call recognition methods, device and storage medium | |
| CN106294508A (en) | A kind of brush amount tool detection method and device | |
| US20170372069A1 (en) | Information processing method and server, and computer storage medium | |
| CN110401780A (en) | A kind of method and device identifying fraudulent call | |
| CN109951609A (en) | A kind of malicious call number processing method and device | |
| CN111783743A (en) | Image clustering method and device | |
| CN109726756A (en) | Image processing method, device, electronic equipment and storage medium | |
| CN109086829A (en) | A kind of method and device that social population administers | |
| CN108712269A (en) | The method for retrieving and device of off-network user | |
| CN109800571A (en) | Event-handling method and device and storage medium and electronic device | |
| CN109905524A (en) | Telephone number recognition methods, device, computer equipment and computer storage medium | |
| CN105224936A (en) | A kind of iris feature information extracting method and device | |
| CN114783021A (en) | Intelligent detection method, device, equipment and medium for wearing of mask | |
| CN109756467A (en) | A kind of recognition methods of fishing website and device | |
| CN109661671A (en) | Improvement using boundary bitmap to image classification | |
| CN110134810A (en) | Retrieve the method and device of image | |
| CN106682058A (en) | Screening method, device and system of application programs | |
| CN117193992A (en) | Model training method, task scheduling device and computer storage medium | |
| CN107766149A (en) | A kind of method and device of the ZooKeeper cluster configurations based on DUBBO |
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 |