CN109918544A - Occupational crime social relation network intelligent analysis method and system based on rough set - Google Patents
Occupational crime social relation network intelligent analysis method and system based on rough set Download PDFInfo
- Publication number
- CN109918544A CN109918544A CN201910191833.3A CN201910191833A CN109918544A CN 109918544 A CN109918544 A CN 109918544A CN 201910191833 A CN201910191833 A CN 201910191833A CN 109918544 A CN109918544 A CN 109918544A
- Authority
- CN
- China
- Prior art keywords
- case
- relationship
- server
- data
- social
- 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 invention belongs to occupational crime fields, it is related to social relation network analysis, specially a kind of occupational crime social relation network intelligent analysis method and system based on rough set, the method includes being classified by the social relation network to middle criminal of having wound up the case, and intellectual analysis is carried out to social relation network based on rough set, it was found that the clue wherein to play a crucial role to case, and remove unrelated clue therein；The data to work to case analysis are filtered out from a variety of data using rough set method based on close case data, are removed to the inoperative redundant data of case analysis, and importance ranking is carried out to various data sources, to provide auxiliary actually to handle a case.
Description
Technical field
The invention belongs to big data analysis technical fields, are related to social relation network analysis, specially a kind of based on coarse
The occupational crime social relation network intelligent analysis method and system of collection.
Background technique
In post Criminal Investigation, the social relation network of suspect is the key that case analysis.In actual case
In part, the social relation network of suspect is generally very huge and complicated, including relatives, classmate, work etc., and these
The building needs of relationship are excavated from multiple data sources, such as Chinese patent CN103714126B is proposed in a kind of push book
The method and device of nationality reading service can effectively excavate data, to effectively teenager be pushed to grow up in study
Reading skills and reading interest are improved in journey.How the social relation network of suspect is constructed from multiple data sources,
And therefrom discovery crime clue is the major issue of occupational crime investigation, and the key point solved a case.Existing occupational crime
According to the experience of handling a case, selection is started with from a certain social relationships, and transfers number from relevant unit for investigation, usually practice expert
According to constructing the social relation network of suspect by artificial mode.For example, main 4 class of certain case suspect
Social relation network is as shown in Figure 1, include four classifications, work class, treats class and friend's class with courtesy at network communication class, wherein work
Make the corresponding social relation network of classification as shown in Fig. 2, the Peer Relationships of the suspect, leader and affiliate etc. are closed
System is divided；
Existing case analysis method depends critically upon the experience of handling a case of practice expert, and basic rely on manually is divided
Analysis, many times will cause case analysis repeatedly, so that under case analysis efficiency is very low.
Summary of the invention
In view of this, traditional occupational crime investigation mode mainly takes manual type to carry out social relation network analysis,
Its efficiency is very low, also affects investigating and prosecuting for case, this patent by the social relation network to middle criminal of having wound up the case into
Row classification, and intellectual analysis is carried out to social relation network based on rough set, find the line wherein to play a crucial role to case
Rope, and remove unrelated clue therein；Investigation foundation is provided for personnel in charge of the case, greatly shortens and handles a case the time, improves effect of handling a case
Rate.
A kind of occupational crime social relation network intelligent analysis method based on rough set of the invention, including following step
It is rapid:
Step 1) obtains the multi-source data in the occupational crime case wound up the case, and constructs primitive society's relational network
Figure；
Primitive society's relational network figure is abstracted into figure G (V, E) by step 2), by n social relation network G1(V,
E1),...,Gn(V,En), it is converted to a relational system (U, R), wherein GnIndicate n-th of social relation network, EnIndicate n-th
The social relationships of all personnel concerning the case in a social relation network；V=U indicates that all personnel concerning the case, R indicate all case-involving
The social relationships of personnel；R={ r1,r2,...,rnIt is one group of binary crelation on U；riIndicate i-th of social relationships；By phase
The G answeredi(E,Ei) obtain, if side (vk,vj) belong to Ei, then have (vk,vj) belong to ri；
Step 3) is based on rough set, screens to the data in relational system, removes wherein useless to case analysis
Personnel concerning the case is divided into criminal and non-criminal by social relationships, thus by relational system (U, R) constituent relation
Decision system (U, R, D)；And reduction is carried out to it, to form the relationships decision system (U, B, D) after reduction；D indicates case-involving
The tally set of personnel, including criminal label d1With non-criminal label d2；B is expressed as all personnel concerning the case after reduction
Social relationships；
Step 4) calculates the relationship importance of each data in the relationships decision system after reduction, and important according to relationship
Degree sequence, checks crime social relation network to be predicted according to the ranking results.
Further, the acquisition modes of the tally set D of the personnel concerning the case include by the social relationships of all personnel concerning the case
R obtains compositive relation R', R'=r1∩r2∩...∩rn, calculate separately about criminal label d1With non-criminal mark
Sign d2Lower apronsd1 Withd2 , according to rough set, calculate positive domain of the D relative to R: POSR(D)=d1 ∪d2 ；Wherein,
[xi]R'Indicate all and xiMeet the set of the object of equivalence relation R', as includes xiR' equivalence class；xi
Indicate the personnel concerning the case in U.
Further, the social relationships B of all personnel concerning the case after reduction described in step 3) meets the following conditions, then B
It is a reduction of set of relationship R；To form the relationships decision system after reduction；
First condition: positive domain of the tally set of personnel concerning the case relative to its subset B, it is opposite with the tally set of personnel concerning the case
In the positive domain of binary crelation R be equal；That is POSB(D)=POSR(D)；
Second condition: having relationship arbitrary in subset B, the tally set of personnel concerning the case about subset B positive domain not
Equal to the positive domain after deleting the relationship in B；I.e.
Further, the relationship weight of each data in the relationships decision system (U, B, D) after reduction described in step 4)
The calculation formula for the property wanted includes that relationship r ∈ B-C, r is indicated relative to the importance of C are as follows: sigr(B, C, D)=γC+{r}
(D)-γC(D), γC+{r}(D) the tally set D of the personnel concerning the case dependence common relative to relationship subset C and relationship r is indicated
Degree；γC(D) dependency degree of the tally set D of personnel concerning the case relative to relationship subset C is indicated.
Further, the tally set D of the personnel concerning the case is expressed as relative to the dependency degree of relationship subset CPOSC(D) positive domain of the tally set D of personnel concerning the case relative to relationship subset C is indicated；It is described to relate to
The dependency degree tally set D of case personnel common relative to relationship subset C and relationship r is expressed asPOSC+{r}(D) indicate personnel concerning the case tally set D relative to close relationship subset C with
And the positive domain that relationship r is common.
It is understood that this patent solves the problems, such as it is to be directed to occupational crime field, and occupational crime refers to: state
Organ, family, state-owned firm, enterprise and institution, people's organization staff corrupt, bribe, dance of acting unfairly from selfish motives using the existing authority of office
Disadvantage is abused one's power, is neglected one's duties, and invades citizen's personal right, democratic rights, destroys the national regulations specification to official activity,
It should give the crime of criminal penalty according to criminal law, including " crime of embezzlement and bribery ", " malfeasance " as defined in the "Criminal Law" and return home
Office clerks takes advantage of one's position and power the infringement citizen personal right of implementation, democratic rights crime.So as long as meeting above-mentioned
The range of crime, be the analyst coverage in this patent.
Based on a kind of occupational crime social relation network intelligent analysis method based on rough set of the invention, the present invention
It also proposed its corresponding system, i.e., a kind of occupational crime social relation network intelligent analysis system based on rough set；It is described
System includes the network communication circle server being connected with occupational crime relational network data source screening server, dinner party circle clothes
Business device, building ring server, circle of friends server；The network communication circle server is logical for obtaining suspect's network
Data in letter；The dinner party circle server is used to obtain the data in suspect's dinner party；The building ring server is used
Data in acquisition suspect's work；The circle of friends server is used to obtain the number in suspect's circle of friends
According to；The Various types of data that the occupational crime relational network data source screening server is used to will acquire is summarized, and logarithm
It is screened according to source, obtains the data source combination of contribution degree maximum (importance highest).
Further, the network communication circle server includes the first statistical module being electrically connected, the first analysis mould
Block, first communication module and first processing module；First statistical module is used to count the call pair of suspect
As, talk times and the duration of call, first analysis module is used to converse with it by talk times and the duration of call
Object carries out corresponding synthesis, and is sent to first processing module；The first processing module is used to for talk times being less than 3 times
And conversation object of the duration of call less than ten minutes is deleted；Treated by processing module for the first communication module
Not deleted communicating data is transmitted to the screening server.
Further, the dinner party circle server includes the second statistical module, second communication module and second processing mould
Block, second statistical module is used to count and suspect carries out the object of recreation, and records its number of activities,
And number of activities are passed to Second processing module, number of activities are subjected to summation analysis, number of activities are greater than to 3 objects
It is passed to second communication module, is passed in the screening server by second communication module.
Further, the building ring server includes third statistical module and third communication module, the third system
Meter module is used to count the object for having work relationship with suspect, is then passed to third communication module, third communication mould
All objects are all passed to the screening server by block.
Further, the circle of friends server includes the 4th statistical module and fourth communication module, the 4th system
Meter module is used to count and suspect is the object of friends, all objects is all sent to, fourth communication mould
All objects are all passed to the screening server by block.
Beneficial effects of the present invention:
The present invention is classified by the social relation network to middle criminal of having wound up the case, and is intelligently divided using big data
Analysis method, constructs different classes of social relation network from multiple data sources, and based on rough set to social relation network into
Row intellectual analysis finds the clue wherein to play a crucial role to case, and removes unrelated clue therein, mentions for personnel in charge of the case
For auxiliary；And traditional occupational crime investigation is not only wasted time, but also be easy to act rashly and alert the enemy so mainly based on confession,
It is unfavorable for the detection of case.The present invention can remove the useless clue in occupational crime investigation.This method innovation tradition is done
The thinking of case method provides investigation foundation for personnel in charge of the case, reduces the workload of personnel in charge of the case, when greatly shortening is handled a case
Between, improve the efficiency handled a case.
Detailed description of the invention
Fig. 1 is the four class social relation network figures of the suspect used in the prior art；
Fig. 2 is the work relationship figure of the suspect used in the prior art；
Fig. 3 is the work relationship figure of the suspect of the use taken out in the prior art；
Fig. 4 is the method flow diagram that the present invention uses；
Fig. 5 is the flow chart of data source screening in the present invention；
Fig. 6 is the flow chart of data source sequence in the present invention；
The social relation network figure that Fig. 7 is suspect A in preferred embodiment in the present invention；
The detailed social relation network figure that Fig. 8 is suspect A in preferred embodiment in the present invention；
The detailed sociogram that Fig. 9 is suspect A in preferred embodiment in the present invention；
Figure 10 is a kind of occupational crime social relation network intelligent analysis system based on rough set of the invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to of the invention real
The technical solution applied in example is clearly and completely described, it is clear that described embodiment is only that present invention a part is real
Example is applied, instead of all the embodiments.
Embodiment 1
Procuratorial organ has had accumulated data of largely handling a case by information-based process.Therefore, this patent is based on
Number of packages evidence of winding up the case filters out the data to work to case analysis using rough set method from a variety of data, removes to case
Part analyzes inoperative redundant data, and carries out importance ranking to various data sources, thus for actually handle a case provide it is auxiliary
It helps.Its main process is as follows:
1, social relation network constructs
For a specific occupational crime case, the data based on separate sources can construct corresponding society and close
It is network.Every one kind community network can be abstracted into a figure G (V, E), and wherein vertex set V indicates all personnel concerning the case, side
Collection E indicates the social relationships of all personnel concerning the case, i.e., if there are corresponding social relationships between two personnel concerning the case, this
There are sides between the corresponding vertex of two personnel concerning the case, and side is otherwise not present.As shown in figure 3, the figure is the social relationships by Fig. 2
The figure that network abstraction obtains.
2, the expression of social relation network
For the ease of the application of rough set method, need to be indicated the relational graph of building the conversion of form.Assuming that one
A case has n social relation network G1(V,E1),...,Gn(V,En), then it can be converted to a relational system (U, R),
Middle U=V, R={ r1,r2,...,rnIt is one group of binary crelation on U.Also, riBy corresponding Gi(E,Ei) obtain, if side
(vk,vj) belong to Ei, then have (vk,vj) belong to ri.For example, being the binary crelation obtained by Fig. 3 as follows:
r1={ (A, B), (B, A), (A, C), (C, A), (A, D), (D, A), (A, E), (E,), A (A, F), (F, A), (A
G),(G,A),(A,H),(H,A),(B,C),(C,B), (E,C),(C,E),(E,F),(F,E),(H,F),(F,H)}
3, data source is screened
In actual case, the social relationships that suspect is related to are many kinds of, and wherein only part is to case
Analysis is worked, and herein by rough set is based on, removes social relationships wherein useless to case analysis.If D is by case
Personnel in part are divided into two classes, and one kind is criminal d1, another kind of is non-criminal d2, then (U, R, D) constitutes a pass
It is decision system.Wherein R={ r1,r2,...,rn, then by R available compositive relation R', R'=r1∩r2∩...∩rn,
Thus, it is possible to calculate separately about d1And d2Lower aprons:
Then according to rough set, positive domain of the D relative to R:
POSR(D)=d1 ∪d2 ；
Here, D is the collection that suspect can be determined whether it is according to known relation relative to the positive domain representation of R
It closes.According to rough set, if removing the part relations in system, and do not change the positive domain of system, then illustrate these relationships for
The analysis of case is redundancy.Therefore, under the premise of not changing system decision-making ability, relationships decision system can be carried out
Reduction.If relationship subset B meets following two condition, claiming B is a reduction of set of relationship R.
POSB(D)=POSR(D)
(1)
Alternatively, according to the condition of above-mentioned setting, following Algorithm for Reduction can be designed:
Algorithm 1: relationships decision system Algorithm for Reduction
Input: relationships decision system f=(U, R, D)
Output: the relationships decision system f=(U, B, D) after reduction
According to above-mentioned algorithm, forIf POSR(D)=POSR-{r}(D), then illustrate relative to decision f'=
For (U, B, D), relationship r is redundancy, then deletes r from decision system f.Repeat aforesaid operations, then can be final
To the reduction of a relationships decision system.
Each of above-mentioned relation decision system binary crelation is practical to correspond to a kind of social relationships, i.e., a kind of data
Source.Therefore, above-mentioned reduction process is actually the screening of data source, and obtained result is exactly the final result screened, i.e. society
It can relationship.Also, obtained result (social relationships) and the result (social relationships) of non-reduction have identical decision-making capability.Cause
For the processing of reduction, so that it is few when in relationship quantity than non-reduction, thus reduce the difficulty of problem.In other words, it handles a case
Personnel are when carrying out case investigation, it is not necessary to comb one by one to all relationships of suspect, as long as and to portion therein
Point relationship carry out combing can be obtained it is identical as a result, so as to improve the efficiency handled a case.
4, data source sorts
Personnel in charge of the case generally requires the path for determining case analysis, i.e., how from a variety of social relationships in actually handling a case
The entrance of middle selection case analysis, and gradually deeply carry out the analysis of case.Here the definition of relationship importance will be provided, it is right
Data source is ranked up.For the relationships decision system after reduction.
A relationship subset in f'=(U, B, D)The tally set D of its personnel concerning the case is relative to relationship subset C
Dependency degree be defined as follows:
Further, relationship r ∈ a B-C, r are defined as follows relative to the importance of C:
sigr(B, C, D)=γC+{r}(D)-γC(D)
Alternatively, according to above-mentioned definition, the sort method of relationship different degree, algorithm be can be designed that
It is as follows:
Algorithm 2: relationship importance sorting
Input: relationships decision system f=(U, B, D)
Output: ranking results C
The practical importance for reflecting each data source of this ranking results.Based on this as a result, personnel in charge of the case can
Preferentially to be checked from these networks, so that the investigation for personnel in charge of the case provides foundation, the time is saved.
As shown in figure 4, the present embodiment step 1 is data preparation stage, carries out the collection of data and to form its original
Network structure；The stage is by using network communication circle server, dinner party circle server, building ring server, circle of friends clothes
The data that business device obtains, i.e. multi-source data；And it is sent in the screening server.Step 2 is to carry out net to the data of collection
Network is constructed to be adapted to be calculated in rough set.Step 3 is to carry out the screening of data source, selects energy by screening server
Determine the smallest collection of network of suspect.Step 4 is ranked up to the data source after screening, and personnel in charge of the case is given
Auxiliary is provided.Wherein, step 3 can be described in further detail as flow chart as shown in Figure 5, all relate to including enable after reduction
The social relationships B=R of case personnel chooses relationship r from relationship R；Judge POSB(D) whether it is equal to POSB-{r}(D), if it is equal
B=B- { r } is then enabled, otherwise directly there is R=R- { r }；And judge whether R is sky, if it is empty, then return result to tally set
In conjunction；It is as shown in Figure 6 that step 4, which can be described in further detail: enabling relationship subsetIt chooses from B-C so that sigr
(B, C, D) maximum relationship r, i.e., until r=max (sigr(B,C,D))；It enables C=C ∪ { r }, B=B- { r }；Until
Embodiment 2
The present embodiment combines specific data, it is assumed that a certain close case analyzed, it is main involved in the case
Suspect is certain leader A in the city W, its community network is analyzed and summarized, obtains the social relation network such as Fig. 7, will be every
A community network source obtains the detailed social relation network such as Fig. 8 after being refined, four relational graphs in Fig. 8 are past from a left side
The right side is successively denoted as (a), (b), (c) and (d) from top to bottom；Fig. 8 (a) indicates the detailed society of the building ring of suspect A
Relationship, Fig. 8 (b) indicate the detailed social relationships of the dinner party circle of suspect A, and Fig. 8 (c) indicates the network of suspect A
The detailed social relationships of circle are communicated, Fig. 8 (d) indicates that the detailed social relationships of the circle of friends of people A are disliked in crime.It again will be each detailed
Social relation network be converted into figure, obtain the form such as Fig. 9, four relational graphs in Fig. 9 from left to right, from top to bottom
It is corresponding to be denoted as (a), (b), (c) and (d)；Fig. 9 (a) is the corresponding transition diagram of Fig. 8 (a), and Fig. 9 (b) is the corresponding conversion of Fig. 8 (b)
Figure, Fig. 9 (c) is the corresponding transition diagram of Fig. 8 (c), and Fig. 9 (d) is the corresponding transition diagram of Fig. 8 (d).
In the case, suspect is 1 and 5, by above-mentioned arrangement it is found that R={ r1,r2,r3,r4, U=1,2,
3,4,5,6 }, and U/D={ (1,5), (2,3,4,6) }, by further arranging, by the binary crelation of each network write as
Lower form:
r1={ (1,2), (2,1), (1,3), (3,1), (1,5), (5,1), (1,4), (4,1), (2,3), (3,2), (2
4),(4,2),(4,6),(6,4)}；
r2={ (1,2), (2,1), (1,5), (5,1), (5,3), (3,5), (5,6), (6,5), (5,4), (4,5) }；
r3={ (1,4), (4,1), (2,5), (5,2), (2,6), (6,2), (3,6), (6,3) }；
r4={ (1,6), (6,1), (2,3), (3,2), (2,5), (5,2), (3,5), (5,3), (3,4), (4,3) }；
In order to meet the reflexive relation in rough set, then:
r'1=U × U-r1=(1,1), (2,2), (3,3), (4,4), (5,5), (6,6), (5,3), (3,5), (5,4),
(4,5),(2,6),(6,2),(1,6),(6,1),(2,5),(5,2),(5,6),(6,5), (3,6),(6,3),(3,4),(4,
3)}
r'2=U × U-r2=(1,1), (2,2), (3,3), (4,4), (5,5), (6,6), (1,3), (3,1), (1,4),
(4,1),(1,6),(6,1),(2,3),(3,2),(6,4),(4,6),(2,6),(6,2), (2,5),(5,2),(3,6),(6,
3),(3,4),(4,3),(2,4),(4,2)}
r'3=U × U-r3=(1,1), (2,2), (3,3), (4,4), (5,5), (6,6), (1,2), (2,1), (1,3),
(3,1),(1,5),(5,1),(1,6),(6,1),(2,3),(3,2),(3,5),(5,3), (5,4),(4,5),(4,6),(6,
4),(5,6),(6,5),(3,4),(4,3), (2,4),(4,2)}
r'4=U × U-r4=(1,1), (2,2) (3,3), (4,4), (5,5), (6,6), (1,2), (2,1), (1,3),
(3,1),(1,4),(4,1),(1,5),(5,1),(5,4),(4,5),(6,4),(4,6), (2,6),(6,2),(5,6),(6,
5),(3,6),(6,3),(2,4),(4,2)}
Then r'=r1'∩r2'∩r3'∩r4'={ (1,1), (2,2), (3,3), (4,4), (5,5), (6,6) }；
The equivalence class for calculating each object in U has:
[1]R={ 1 }, [2]R={ 2 }, [3]R={ 3 }, [4]R={ 4 }, [5]R={ 5 }, [6]R={ 6 }
Then the tally set D of personnel concerning the case is as follows relative to the positive domain of the social relationships R of all personnel concerning the case:
POSR(D)={ 1,2,3,4,5,6 }；
Then by r1This attribute is deleted, and the positive domain for calculating remaining attribute obtains:
BecauseSo attribute r1It is necessary, is irreducible.Then to surplus
Under attribute successively carry out single deletion then repeat aforesaid operations, find r3It is reducible, and r2And r4It is irreducible.
Therefore, the attribute set after final reduction is { r1,r2,r4}.By above-mentioned analysis it is found that r1、 r2And r4What is respectively represented is
" building ring ", " dinner party circle " and " circle of friends ".That is, the case is only to need to carry out in these three circles in investigation
Investigation.
Then, the sequence of data source is carried out to the decision system after above-mentioned reduction.Firstly, for the system f' after reduction
=(U, B, D), the set of relations B={ r after reduction1,r2,r4, thenCalculating its dependency degree has: r1=0, r2=
0.17, r4=0, then it is r that dependency degree is maximum at this time2, then again to remaining ri∈ B-C, calculating its different degree has:Then the ranking results of final data source are (descending arrangement): r2, r4, r1.So
In the case investigation of later same type, personnel in charge of the case can preferentially carry out investigations from dinner party circle, because of suspicion of crime Genus Homo
It is very big in the circle a possibility that, then successively checked down again.So as to save the time, improve efficiency, to the people that handles a case
Member provides investigation direction.
Further, occupational crime social relation network intelligent analysis method mentioned above can be further expanded,
Propose its corresponding system, a kind of occupational crime social relation network intelligence based on rough set specially as shown in Figure 10
It can analysis system.
The system comprises the network communication circle services being connected with occupational crime relational network data source screening server
Device, dinner party circle server, building ring server, circle of friends server；The network communication circle server is disliked for obtaining crime
Doubt the data in people's network communication；The dinner party circle server is used to obtain the data in suspect's dinner party；The work
Circle server is used to obtain the data in suspect's work；The circle of friends server is for obtaining friend suspect
Data in friend's circle；The Various types of data that the occupational crime relational network data source screening server is used to will acquire carries out
Summarize, and data source is screened, obtains the data source combination of contribution degree maximum (importance highest).
The network communication circle server includes the first statistical module being electrically connected, the first analysis module, the first communication
Module and first processing module；First statistical module be used to count conversation object, the talk times of suspect with
And the duration of call, first analysis module are used to carry out with its conversation object by talk times and the duration of call corresponding
It is comprehensive, and it is sent to first processing module；The first processing module is used to talk times being less than 3 times and the duration of call
Conversation object less than ten minutes is deleted；The first communication module is by processing module treated not deleted call
Data are transmitted to the screening server.
Dinner party circle server includes the second statistical module, second communication module and Second processing module, and described the
Two statistical modules are used to count and suspect carries out the object of recreation, and record its number of activities, and by movable time
Number of activities are carried out summation analysis, the object by number of activities greater than 3 times is passed to second by the incoming Second processing module of number
Communication module is passed in the screening server by second communication module.
The building ring server includes third statistical module and third communication module, and the third statistical module is used
There is the object of work relationship in statistics and suspect, is then passed to third communication module, third communication module will own
Object be all passed to the screening server.
The circle of friends server includes the 4th statistical module and fourth communication module, and the 4th statistical module is used
In the object that statistics and suspect are friends, all objects are all sent to, fourth communication module will own
Object be all passed to the screening server.
If above-mentioned described example case applies in the system, can be described as follows: in system shown in Fig. 10
In, the data about each social source of suspect are obtained by corresponding server respectively, such as " network communication circle service
Device " obtains the related data in suspect's network communication, and so on.Then the related data that will acquire is aggregated into
In intermediate " screening of occupational crime relational network data source " server, then use occupational crime social relationships mentioned above
Network intelligence analysis method carries out data source screening, obtains the maximum data source combination of contribution degree, mentions for the detection of subsequent case
For direction.
In general case, " screening of the occupational crime relational network data source " server at center be can connect in case
Required other servers carry out the acquisition of data source, then again through the invention in the analysis method mentioned carry out data
Source screening, finally obtains a result.
To avoid repeating, the correlated characteristic of the method and system in the present invention can be quoted mutually.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include: ROM, RAM, disk or CD etc..
In addition, term " first ", " second ", " third ", " the 4th " are used for description purposes only, and should not be understood as indicating
Or imply relative importance or implicitly indicate the quantity of indicated technical characteristic, define as a result, " first ", " second ",
The feature of " third ", " the 4th " can explicitly or implicitly include at least one of the features.
Embodiment provided above has carried out further detailed description to the object, technical solutions and advantages of the present invention,
It should be understood that embodiment provided above is only the preferred embodiment of the present invention, it is not intended to limit the invention,
Any modification, equivalent substitution, improvement and etc. made for the present invention all within the spirits and principles of the present invention, should be included in
Within protection scope of the present invention.
Claims (10)
1. a kind of occupational crime social relation network intelligent analysis method based on rough set, which is characterized in that the method packet
Include following steps:
Step 1) obtains the multi-source data in the occupational crime case wound up the case, and constructs primitive society's relational network figure；
Primitive society's relational network figure is abstracted into figure G (V, E) by step 2), by n social relation network G1(V,E1),...,Gn
(V,En), it is converted to a relational system (U, R), wherein GnIndicate n-th of social relation network, EnIndicate n-th of social relationships
The social relationships of all personnel concerning the case in network；V=U indicates that all personnel concerning the case, R indicate the society of all personnel concerning the case
Relationship；R={ r1,r2,...,rnIt is one group of binary crelation on U；riIndicate i-th of social relationships；
Step 3) is based on rough set, screens to the data in relational system, removes society wherein useless to case analysis
Personnel concerning the case is divided into criminal and non-criminal by relationship, thus by relational system (U, R) constituent relation decision system
(U,R,D)；And reduction is carried out to it, to form the relationships decision system (U, B, D) after reduction；The mark of D expression personnel concerning the case
Label collection, including criminal label d1With non-criminal label d2；The society that B is expressed as all personnel concerning the case after reduction closes
System；
Step 4) calculates the relationship importance of each data in the relationships decision system after reduction, and arranges according to relationship different degree
Sequence checks crime social relation network to be predicted according to the ranking results.
2. a kind of occupational crime social relation network intelligent analysis method based on rough set according to claim 1,
It is characterized in that, the acquisition modes of the tally set D of the personnel concerning the case include being synthesized by the social relationships R of all personnel concerning the case
Relationship R', R'=r1∩r2∩...∩rn, calculate separately about criminal label d1With non-criminal label d2Lower apronsd1 Withd2 , according to rough set, calculate positive domain of the D relative to R: POSR(D)=d1 ∪d2 ；Wherein,
[xi]R'Indicate all and xiMeet the set of the object of equivalence relation R', as includes xiR' equivalence class；xiIndicate U
In personnel concerning the case.
3. a kind of occupational crime social relation network intelligent analysis method based on rough set according to claim 1,
It is characterized in that, the social relationships B of all personnel concerning the case after reduction described in step 3) meets the following conditions, then B is set of relations
Close a reduction of R；To form the relationships decision system after reduction；
First condition: positive domain of the tally set of personnel concerning the case relative to its subset B, the tally set with personnel concerning the case is relative to binary
The positive domain of relationship R is equal；That is POSB(D)=POSR(D)；
Second condition: have for relationship arbitrary in subset B, the tally set of personnel concerning the case is not equal to about the positive domain of subset B
Positive domain after deleting the relationship in B；I.e.
4. a kind of occupational crime social relation network intelligent analysis method based on rough set according to claim 1,
It is characterized in that, the meter of the relationship importance of each data in relationships decision system (U, B, D) after reduction described in step 4)
Calculating formula includes that relationship r ∈ B-C, r is indicated relative to the importance of C are as follows: sigr(B, C, D)=γC+{r}(D)-γC(D),
γC+{r}(D) the tally set D of the personnel concerning the case dependency degree common relative to relationship subset C and relationship r is indicated；γC(D) it indicates
Dependency degree of the tally set D of personnel concerning the case relative to relationship subset C.
5. a kind of occupational crime social relation network intelligent analysis method based on rough set according to claim 4,
It is characterized in that, the tally set D of the personnel concerning the case is expressed as relative to the dependency degree of relationship subset CPOSC(D) positive domain of the tally set D of personnel concerning the case relative to relationship subset C is indicated；It is described to relate to
The dependency degree tally set D of case personnel common relative to relationship subset C and relationship r is expressed asPOSC+{r}(D) indicate personnel concerning the case tally set D relative to close relationship subset C with
And the positive domain that relationship r is common.
6. a kind of occupational crime social relation network intelligent analysis system based on rough set, which is characterized in that the system packet
Include the network communication circle server being connected with occupational crime relational network data source screening server, dinner party circle server, work
Make circle server, circle of friends server；The network communication circle server is used to obtain the number in suspect's network communication
According to；The dinner party circle server is used to obtain the data in suspect's dinner party；The building ring server is used for the criminal of acquisition
Suspect doubts the data in artificial make；The circle of friends server is used to obtain the data in suspect's circle of friends；It is reported
The Various types of data that business crime relational network data source screening server is used to will acquire is summarized, and is sieved to data source
Choosing obtains the highest data source combination of importance.
7. a kind of occupational crime social relation network intelligent analysis system based on rough set according to claim 6,
It is characterized in that, the network communication circle server includes the first statistical module being electrically connected, the first analysis module, the first communication
Module and first processing module；First statistical module be used to count conversation object, the talk times of suspect with
And the duration of call, first analysis module are used to carry out with its conversation object by talk times and the duration of call corresponding comprehensive
It closes, and is sent to first processing module；The first processing module is used to for talk times being less than 3 times and the duration of call is less than
Ten minutes conversation objects are deleted；Treated that not deleted communicating data passes by processing module for the first communication module
Transport to the screening server.
8. a kind of occupational crime social relation network intelligent analysis system based on rough set according to claim 6,
It being characterized in that, dinner party circle server includes the second statistical module, second communication module and Second processing module, and described the
Two statistical modules are used to count and suspect carries out the object of recreation, and record its number of activities, and by movable time
Number of activities are carried out summation analysis, the object by number of activities greater than 3 times is passed to second and leads to by the incoming Second processing module of number
Believe module, is passed in the screening server by second communication module.
9. a kind of occupational crime social relation network intelligent analysis system based on rough set according to claim 6,
It is characterized in that, the building ring server includes third statistical module and third communication module, and the third statistical module is used
There is the object of work relationship in statistics and suspect, is then passed to third communication module, third communication module will own
Object be all passed to the screening server.
10. a kind of occupational crime social relation network intelligent analysis system based on rough set according to claim 6,
It is characterized in that, the circle of friends server includes the 4th statistical module and fourth communication module, and the 4th statistical module is used
In the object that statistics and suspect are friends, all objects are all sent to, fourth communication module will own
Object be all passed to the screening server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910191833.3A CN109918544B (en) | 2019-03-14 | 2019-03-14 | Rough set-based social relationship network intelligent analysis method and system for job crime |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910191833.3A CN109918544B (en) | 2019-03-14 | 2019-03-14 | Rough set-based social relationship network intelligent analysis method and system for job crime |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109918544A true CN109918544A (en) | 2019-06-21 |
| CN109918544B CN109918544B (en) | 2021-03-12 |
Family
ID=66964766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910191833.3A Active CN109918544B (en) | 2019-03-14 | 2019-03-14 | Rough set-based social relationship network intelligent analysis method and system for job crime |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109918544B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532317A (en) * | 2019-08-26 | 2019-12-03 | 北京百度网讯科技有限公司 | Case analysis data processing method, device, equipment and readable storage medium storing program for executing |
| CN112016004A (en) * | 2020-08-21 | 2020-12-01 | 重庆邮电大学 | Multi-granularity information fusion-based job crime screening system and method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103052130A (en) * | 2012-11-20 | 2013-04-17 | 南京邮电大学 | Rough-set-based data fusion method for wireless multimedia sensor network |
| CN104408149A (en) * | 2014-12-04 | 2015-03-11 | 威海北洋电气集团股份有限公司 | Criminal suspect mining association method and system based on social network analysis |
| CN104915430A (en) * | 2015-06-15 | 2015-09-16 | 南京邮电大学 | Method for obtaining constraint relation rough set rules based on MapReduce |
| CN105844414A (en) * | 2016-03-25 | 2016-08-10 | 北京石油化工学院 | Method for evaluating dangerous chemical safety management |
| CN106126523A (en) * | 2016-06-12 | 2016-11-16 | 中国科学院软件研究所 | A kind of counterfeit money Crime Information analyzes system and the method for analysis |
| US20180130124A1 (en) * | 2016-11-10 | 2018-05-10 | National Tsing Hua University | Method for product service system classification and service transformation |
| CN108537259A (en) * | 2018-03-27 | 2018-09-14 | 北京交通大学 | Train control on board equipment failure modes and recognition methods based on Rough Sets Neural Networks model |
-
2019
- 2019-03-14 CN CN201910191833.3A patent/CN109918544B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103052130A (en) * | 2012-11-20 | 2013-04-17 | 南京邮电大学 | Rough-set-based data fusion method for wireless multimedia sensor network |
| CN104408149A (en) * | 2014-12-04 | 2015-03-11 | 威海北洋电气集团股份有限公司 | Criminal suspect mining association method and system based on social network analysis |
| CN104915430A (en) * | 2015-06-15 | 2015-09-16 | 南京邮电大学 | Method for obtaining constraint relation rough set rules based on MapReduce |
| CN105844414A (en) * | 2016-03-25 | 2016-08-10 | 北京石油化工学院 | Method for evaluating dangerous chemical safety management |
| CN106126523A (en) * | 2016-06-12 | 2016-11-16 | 中国科学院软件研究所 | A kind of counterfeit money Crime Information analyzes system and the method for analysis |
| US20180130124A1 (en) * | 2016-11-10 | 2018-05-10 | National Tsing Hua University | Method for product service system classification and service transformation |
| CN108537259A (en) * | 2018-03-27 | 2018-09-14 | 北京交通大学 | Train control on board equipment failure modes and recognition methods based on Rough Sets Neural Networks model |
Non-Patent Citations (5)
| Title |
|---|
| ADNAN AMIN: ""Classification of Cyber Attacks based on Rough Set Theory"", 《2015 FIRST INTERNATIONAL CONFERENCE ON ANTI-CYBERCRIME (ICACC)》 * |
| 张盼盼: ""粗糙集在犯罪信息中的应用"", 《智能计算机与应用》 * |
| 王钢 等: ""犯罪团伙网络关系模型及分析方法"", 《中国人民公安大学学报（社会科学版）》 * |
| 罗艺 等: ""Rough集和数据挖掘应用于案件综合分析"", 《微计算机信息》 * |
| 聂冰 等: ""基于粗糙集属性重要性的数据约简"", 《电子测量技术》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110532317A (en) * | 2019-08-26 | 2019-12-03 | 北京百度网讯科技有限公司 | Case analysis data processing method, device, equipment and readable storage medium storing program for executing |
| CN110532317B (en) * | 2019-08-26 | 2022-06-14 | 北京百度网讯科技有限公司 | Case analysis data processing method, device, equipment and readable storage medium |
| CN112016004A (en) * | 2020-08-21 | 2020-12-01 | 重庆邮电大学 | Multi-granularity information fusion-based job crime screening system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109918544B (en) | 2021-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106550155B (en) | Swindle sample is carried out to suspicious number and screens the method and system sorted out and intercepted | |
| CN106780263B (en) | High-risk personnel analysis and identification method based on big data platform | |
| CN110310089A (en) | A kind of meeting room management system and method based on intelligent dispatching algorithm | |
| CN109766359B (en) | Credit subject comprehensive analysis management system and method | |
| CN106778876A (en) | User classification method and system based on mobile subscriber track similitude | |
| CN109918544A (en) | Occupational crime social relation network intelligent analysis method and system based on rough set | |
| CN109948692A (en) | Picture detection method is generated based on the computer of multiple color spaces convolutional neural networks and random forest | |
| WO2022057108A1 (en) | Federated-learning-based personal qualification evaluation method, apparatus and system, and storage medium | |
| Zhou et al. | Federated conditional generative adversarial nets imputation method for air quality missing data | |
| CN107527240A (en) | A kind of operator's industry product Praise effect identification system and method | |
| CN112507006A (en) | Power grid enterprise operation data integration system based on cloud | |
| Ma et al. | Meta-weight graph neural network: Push the limits beyond global homophily | |
| CN106789572A (en) | A kind of instant communicating system and instant communication method for realizing self adaptation message screening | |
| CN111918226B (en) | Real-time signaling-based method and device for analyzing international high-settlement embezzlement behavior | |
| CN103984760B (en) | The data structure of content oriented distribution subscription system and its compound event matching process | |
| CN110276949A (en) | Forecasting Approach for Short-term Traffic Flow based on machine learning | |
| McLOUGHLIN et al. | A tale of three tribes: UK MPs, Twitter and the EU Referendum campaign | |
| CN108491481B (en) | Intelligent identification active early warning system and method for party involved in intellectual property right infringement | |
| CN114897517B (en) | Text and travel consumption data management method based on block chain | |
| Calderoni | 29 Transnational Organized Crime Networks | |
| Huang et al. | FVRD: fishing vessel relations discovery through VMS trace analysis | |
| Zhou et al. | Lstm based network flow forecasting in high-speed railway data network | |
| CN109933676A (en) | A kind of information base construction method and device | |
| He et al. | The title of the paper: research on insurance marketing application based on hash link-table improved association rule algorithm | |
| Jun'e | University Financial Management and Decision Analysis based on Decision Tree Algorithm |
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 |