CN108733683A - A kind of method and device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data - Google Patents
A kind of method and device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data Download PDFInfo
- Publication number
- CN108733683A CN108733683A CN201710247747.0A CN201710247747A CN108733683A CN 108733683 A CN108733683 A CN 108733683A CN 201710247747 A CN201710247747 A CN 201710247747A CN 108733683 A CN108733683 A CN 108733683A
- Authority
- CN
- China
- Prior art keywords
- entity object
- data
- event
- occurred
- people
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000000694 effects Effects 0.000 claims description 54
- 230000006399 behavior Effects 0.000 claims description 32
- 230000004927 fusion Effects 0.000 claims description 11
- 238000005201 scrubbing Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 3
- 241000406668 Loxodonta cyclotis Species 0.000 claims description 2
- 238000001228 spectrum Methods 0.000 claims description 2
- 238000013178 mathematical model Methods 0.000 description 15
- 230000002159 abnormal effect Effects 0.000 description 12
- 238000004458 analytical method Methods 0.000 description 10
- 238000001514 detection method Methods 0.000 description 9
- 238000011840 criminal investigation Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000005336 cracking Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000003032 molecular docking Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 230000000241 respiratory effect Effects 0.000 description 2
- 230000035807 sensation Effects 0.000 description 2
- 230000004308 accommodation Effects 0.000 description 1
- 230000003416 augmentation Effects 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Alarm Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of to be sounded out the people in a given scope one by one in order to break a criminal case the method and device of exploration event clue based on data, the method includes：According to the information of entity object, the relation map based on the entity object is built；According to the characteristic information for the event that occurred, determine that being used for handling the data that event has occurred sounds out the people in a given scope one by one in order to break a criminal case condition；It is sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determined and the related entity object of event has occurred with described.The present invention quickly finds clue and determination entity object related with the event that occurred it can be found that the association hidden inside data.
Description
Technical field
The present invention relates to data analysis field, more particularly to the method and dress of a kind of exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data
It sets.
Background technology
With the rapid development of mobile Internet, modern industry and science and technology, network bandwidth is greatly improved, smart mobile phone
Hardware performance constantly promoted, high-speed railway is promoted, and civil aviaton is more and more flourishing, and family saloon is popularized, these variation
Living standards of the people are made constantly to be promoted, life is more and more convenient, and various countries' cultural exchanges are also more and more frequent.Meanwhile it is modern
This scientific and technological double-edged sword also makes the activity of offender more and more convenient, and crime cost is lower and lower, and means are more and more abundant, side
Method is more and more hidden, and to government, relatively backward management level, technical capability and the hardware device shape of public security and criminal investigation system
At sharp contrast, to government, public security and criminal investigation system generate great challenge, are formed and are threatened to people life property safety.
It would therefore be highly desirable to one kind is needed to closely follow era development, it can be to smart mobile phone, railway, civil aviaton, automobile, entry and exit etc.
Content carries out the intelligence analysis system of the technological means monitored based on population information, to assist government, public security and criminal investigation department
It fights crime, quick-fried to fear, people life property safety is safeguarded in the public harm of early warning.
There are disadvantages for traditional intelligence analysis system for cracking of cases：
1. the type of data acquisition channel and gathered data is all more single, a large amount of new types of data is not yet integrated into public security system
In system；
2. Data Management Analysis ability is excessively single, not yet forms effective big data and excavate means, in current information
In processing, conventional data analysis tool and analysis method still occupy leading position, can only simply tire out when handling many data
Add, compare and single-mode calculates, the internal association of information excavates insufficient, cannot fully meet the needs of public security work.
Invention content
The method and device of a kind of exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data provided according to embodiments of the present invention, by dividing
Association hiding inside mass data is analysed, event clue is fast and effeciently explored.
A kind of method of the exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data provided according to embodiments of the present invention, including：
According to the information of entity object, the relation map based on the entity object is built；
According to the characteristic information for the event that occurred, determine that being used for handling the data that event has occurred sounds out the people in a given scope one by one in order to break a criminal case condition；
It is sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determination has occurred with described
The related entity object of event.
Preferably, the information according to the entity object, relation map of the structure based on the entity object include：
Using the essential information of the entity object as keyword, obtain has the other of activity relationship with the entity object
Entity object, and establish the relationship for including other entity objects with activity relationship centered on the entity object
Collection of illustrative plates.
Preferably, described to be sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determine with
The related entity object of event that occurred includes：
Sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, filter out with the relation map that event has occurred is unrelated, obtain with it is described
The related relation map of generation event, and obtain the corresponding entity object of the relation map.
A kind of device of the exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data provided according to embodiments of the present invention, including：
Module is built, for the information according to entity object, builds the relation map based on the entity object；
Setup module determines for the characteristic information according to the event that occurred and is used for handling the number that event has occurred
According to the condition of sounding out the people in a given scope one by one in order to break a criminal case；
It sounds out the people in a given scope one by one in order to break a criminal case module, for being sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, is determined
The related entity object of event has occurred with described.
Preferably, the structure module is specifically used for using the essential information of the entity object as keyword, acquisition and institute
Stating entity object has all other entity object of activity relationship, and it includes described to establish centered on the entity object
The relation map of every other entity object with activity relationship.
Preferably, the module of sounding out the people in a given scope one by one in order to break a criminal case is specifically used for sounding out the people in a given scope one by one in order to break a criminal case condition using the data, filters out and event has occurred with described
Unrelated relation map obtains that the related relation map of event has occurred with described, and obtains the corresponding reality of the relation map
Body object.
According to a kind of device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data that another embodiment of the present invention provides, including：
Memory；And
Processor is coupled to the memory, and for the information according to entity object, structure is based on the entity pair
The relation map of elephant determines that being used for handling the data that event has occurred sounds out the people in a given scope one by one in order to break a criminal case item according to the characteristic information for the event that occurred
Part, and sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determines and event has occurred with described
Related entity object.
Technical solution provided in an embodiment of the present invention has the advantages that：
The embodiment of the present invention quickly finds clue it can be found that the association hidden inside data, and determines and thing has occurred
The related entity object of part can be applied to auxiliary public security department to the detection work of case, quick lock in case suspect.
Description of the drawings
Fig. 1 is the method flow diagram for the exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data that the embodiment of the present invention one provides；
Fig. 2 is that provided by Embodiment 2 of the present invention sounded out the people in a given scope one by one in order to break a criminal case based on data explore the flow chart of case clue；
Fig. 3 is that being sounded out the people in a given scope one by one in order to break a criminal case based on data for the offer of the embodiment of the present invention three explore the flow chart of case clue；
Fig. 4 is the device block diagram for the exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data that the embodiment of the present invention four provides.
Specific implementation mode
Below in conjunction with attached drawing to a preferred embodiment of the present invention will be described in detail, it should be understood that described below is excellent
Select embodiment only for the purpose of illustrating and explaining the present invention and is not intended to limit the present invention.
Embodiment one
The embodiment of the present invention can be applied to government, public security and criminal investigation system, by taking the application scenarios of public security investigation as an example, such as scheme
What 1 embodiment of the present invention one provided is sounded out the people in a given scope one by one in order to break a criminal case based on data shown in the method flow diagram of exploration event clue, and step includes：
Step S101：According to the information of entity object, the relation map based on the entity object is built.
Before step S101, it is necessary first to obtain phone number from different data sources gathered data, such as from operator
And message registration, obtain banking business data from bank.Since data come from different data sources, it is therefore desirable to the data of acquisition
It is pre-processed, specifically, the data progress data scrubbing and data for the separate sources of any one entity object are melted
It closes, to obtain the information of the entity object.
Wherein, entity object can be people, can also be object, such as vehicle.
Wherein, the information of the entity object includes the essential information and action message of entity object.It is with entity object
Artificial example, essential information may include ID card No., phone number, bank's card number, driver license number, and action message may include
Message registration, transaction record, banking record, trip record.
Step 101 includes：Using the essential information of the entity object as keyword, obtains to have with the entity object and live
Other entity objects of dynamic relationship, and establish other realities with activity relationship comprising described in centered on the entity object
The relation map of body object.For example, using the phone number of a certain entity object as keyword, it may be determined that converse with it other
Entity object, thus the social relationships collection of illustrative plates established should be related to the talk activity of the entity object it is associated it is all its
Its entity object.
Step S102：According to the characteristic information for the event that occurred, determine that being used for handling the data that event has occurred touches
Row's condition.
Specifically, step S102 is sounded out the people in a given scope one by one in order to break a criminal case condition so that the characteristic information of event has occurred as data, the characteristic information
It can be time of the event that occurred, place, personage, such as the participation personage of event has occurred and has been determined as 40 years old or more, then
It can will be more than at age 40 years old and sound out the people in a given scope one by one in order to break a criminal case condition as data.
Step S103：It is sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determined and institute
State the related entity object of the event of generation.
Step S103 includes：Sounded out the people in a given scope one by one in order to break a criminal case condition using the data, filter out with the relation map that event has occurred is unrelated,
It obtains that the related relation map of event has occurred with described, and obtains the corresponding entity object of the relation map.For example, by year
Age is more than 40 years old as data condition of sounding out the people in a given scope one by one in order to break a criminal case and filters out four entity objects under 10 years of age by denoising.
The embodiment of the present invention be by the analysis to mass data, based on the association including people, the essential information of vehicle object,
The sociogram (i.e. social relationships collection of illustrative plates) of automatic structure object, is then based on the sociogram, user can be according to tool
The affair character of body, on the sociogram, setting data are sounded out the people in a given scope one by one in order to break a criminal case condition, which is repeated at denoising
Reason, until obtaining and the matched criminal profiling of the affair character, locking case suspect.
The method can also using entity object information, structure the relation map based on the entity object it
Afterwards, the action message according to the entity object within a preset period of time carries out similitude classification to the entity object, and is
The respective crawler behavior model with action message is established per class entity object.In this way, when an event occurs, it can be according to institute
The action message for stating the entity object that the event of generation is related to determines the crawler behavior model belonging to the action message, and root
According to definitive result, the probability that event has occurred described in the entity object participation is predicted.For example, for auxiliary public security department to case
The detection work of part, the crawler behavior model can be divided into normal social activities citizen behavior mathematical model and abnormal hair society
Movable citizen's behavior mathematical model.It, can be according to the action message for the entity object to be selected that case has occurred, really after case occurs
The social activities behavior of the fixed entity object to be selected belongs to normal social activities citizen behavior mathematical model or abnormal hair society
The movable citizen's behavior mathematical model of meeting, if belonging to abnormal hair social activities citizen's behavior mathematical model, according to its activity letter
Breath, predicts the probability that its social activities behavior is suspected of committing a crime, and analyze associated information, helps to lock case suspicion
People.
Embodiment two
Fig. 2 is that provided by Embodiment 2 of the present invention sounded out the people in a given scope one by one in order to break a criminal case based on data explore the flow chart of case clue, by this
Method quickly finds clue and portrays criminal profiling, as shown in Fig. 2, this method it can be found that the association hidden inside data
Include the following steps：
Step S201：Gathered data, including acquisition demographic data, vehicle data, trip data, transaction data, hotel are lived
Place data, communicating data, social data, banking business data acquire multiple data sources data.
Data docking acquisition mode can be connect by data (DB) interface mode, file interface mode or WWW (WEB)
The self defined interface that mouth mode or both sides negotiate docks acquisition mode, carries out data acquisition.
Step S202：Handle data.
Data scrubbing is carried out first, mainly reaches standardized format, abnormal data removing, error correcting, duplicate data
It removes, it is ensured that the referential integrity of data and the accuracy of data mining.
Then data fusion is carried out, using the essential information of people as keyword, essential information includes ID card No., passport No.
Code, phone number, bank's card number, driver license number, social security number get up the separate sources data attribute fusion of same person, warp
Refinement, processing and conversion are crossed, finally by the data integration of dispersion a to tables of data, the data after fusion are packed into basis
Database is uniformly stored in data center.
After data processing, social activities attribute information (the i.e. activity letter of embodiment one of reflection people can be got
Breath), such as transaction data, buy activity data, social activity data, trip data, message registration data.
Step S203：Analyze data.
Using the essential information of entity object as keyword, for example, the case where entity object is behaved, then the essential information of people
Including ID card No., passport number, phone number, bank's card number, driver license number, social security number, in the number that step S202 is established
According to the respiratory sensation analysis carried out on the basis of library centered on the entity, so-called to establish respiratory sensation just refer to the basic of entity
Information is keyword, all information of all other entities that social relationships occur with the entity is analyzed, to establish the reality
Social relationships collection of illustrative plates (the i.e. relational graphs of embodiment one with other all solids that social relationships entity occurs centered on body
Spectrum), that is, collect all social activities characteristic informations that the entity is related to.Further, the above method is recycled,
Within some period, for example, it is 1 year nearest in, acquisition owner's entity object all social activities characteristic informations (implement
The action message of example one).Further, using the clustering algorithm of unsupervised learning proprietary social activities is collected to above-mentioned
Characteristic information carries out the classification by similitude, is classified as several classifications.Wherein, clustering technique is divided into three steps：Data preparation,
Similarity defines and clustering technique.Data preparation is exactly owner's entity object that second step is got after data processing
All social activities characteristic informations.Similarity definition is exactly to define the distance between data, can use a variety of methods, such as：
Minkowski (Minkowski) distance definition, two n dimension variable a (x11, x12 ..., x1n) and b (x21, x22 ..., x2n)
Between Minkowski distance definitions be：Wherein p is a variable element.
Clustering technique can also use K mean values (K-means) clustering, that is, fast clustering analysis, its first logarithm
According to preliminary classification is carried out, then gradually adjusts, obtain final classification." target of K-means algorithms is incited somebody to action according to input parameter k
Data set is divided into k cluster, and algorithm uses the newer method of iteration：In each round, according to k reference point by surrounding point
K cluster is separately constituted, and the barycenter of each cluster will be by the reference point as next round iteration, wherein institute in the barycenter, that is, cluster
Average value a little, that is, geometric center.Iteration makes the reference point chosen become closer to true cluster barycenter, so poly-
Class effect is become better and better.K-means algorithms randomly select K point as initial cluster center first, then calculate each sample
To the distance of cluster centre, class sample being grouped into where that cluster centre nearest from it；New class after adjustment is calculated
New cluster centre illustrates that sample adjustment terminates, clustering criteria function if adjacent cluster centre twice does not have any variation
It has restrained.One feature of this algorithm be will investigate in each iteration each sample classification it is whether correct, if not just
Really it is necessary to being adjusted.After whole samples adjust, then cluster centre is changed, into next iteration.If primary
In iterative algorithm, all samples are correctly classified, then do not have adjustment, and cluster centre there will not be any variation, this mark
It clustering criteria function to have restrained, therefore algorithm terminates.
Based on above-mentioned clustering algorithm make in some period of selection owner's entity object (in 1 year such as nearest) into
Row is sorted out by similitude, such as is classified as 2 classifications, respectively normal social activities citizen and abnormal social activities citizen.Into one
Step establishes normal social activities citizen behavior mathematical model and different to above-mentioned two category set using model-naive Bayesian respectively
Normal social activities citizen's behavior mathematical model.
Step S204：According to the event or case specifically occurred discriminate line is utilized for the suspicious people participated in case
Property regression model by the characteristic information of suspicious people respectively with normal social activities citizen behavior mathematical model and abnormal social activities
Citizen's behavior mathematical model matches, and predicts the probability that the social activities of suspicious people is suspected of committing a crime, explores solution or detection
The information of the needs of the event or case provides help to clear up a criminal case.
Step S205：Data are sounded out the people in a given scope one by one in order to break a criminal case.
On the basis of the social relationships collection of illustrative plates that step S203 is established, made according to the specific characteristic information that event or case occurs
Sound out the people in a given scope one by one in order to break a criminal case condition for data, on social relationships collection of illustrative plates setting sound out the people in a given scope one by one in order to break a criminal case condition, social relationships collection of illustrative plates is carried out according to the condition of setting
The information filtering with the event incoherent people, object is fallen in denoising.By repeatedly being set not to social relationships collection of illustrative plates repeatedly
With data sound out the people in a given scope one by one in order to break a criminal case condition, explore and obtain the social relationships collection of illustrative plates identical with the event or case, obtain the event or case phase
The people of pass, the entity information of object, according to the solution and detection of the information ancillary events or case.
Wherein, the condition sounded out the people in a given scope one by one in order to break a criminal case can be include event occur time, place, the participation of the personage event entity
Characteristic attribute.
Wherein, arbitrarily occur between two entities of social relationships to allow that condition of sounding out the people in a given scope one by one in order to break a criminal case is arranged, can be sounded out the people in a given scope one by one in order to break a criminal case by setting
Condition filter falls and the event or the incoherent entity information of case, that is, denoising.
The embodiment of the present invention can sound out the people in a given scope one by one in order to break a criminal case from mass data case clue and automatically structure with the case coincide society
Relation map assists cracking of cases.
Embodiment three
Fig. 3 is that being sounded out the people in a given scope one by one in order to break a criminal case based on data for the offer of the embodiment of the present invention three explore the flow chart of case clue, utilizes this
Embodiment, detection management of the auxiliary public security department to case, as shown in figure 3, step includes：
Step S301：Gathered data.
In order to which bigger plays the value of system, need to acquire abundant data, including acquisition demographic data, vehicle data,
Trip data, transaction data, hotel accommodations data, communicating data, social data, consumption data, that is, acquire a plurality of types of numbers
According to.
When the system is applied to public security department, in order to play the bigger value of the system, the detection rate of case is improved
And working efficiency, the system for accessing multiple departments is needed, abundant data are acquired, for example, in order to obtain trip data, needs to connect
Enter relevant departments' system including civil aviaton, highway, railway, obtains corresponding trip data.
Data docking acquisition mode can pass through DB interface modes, file interface mode or WEB interface modes or both sides
The self-defined privately owned interface negotiated docks acquisition mode, carries out data acquisition.
Step S302, data are handled.
Cleaning fusion treatment is carried out to the collected data of step S301, identifies that the basis corresponding to these data is real
Body, such as population will be associated about the data of the same entity, and to the data attribute of same entity carry out value compare and
Logical relation compares, and the part of corroborate each other in discovery separate sources data (good data) and conflicting (bad data) passes through
Inconsistent data (i.e. bad data) are issued data source offer unit and repaired by data pipe flow control journey, thus by the same reality
The separate sources data attribute fusion of body is got up, and the model of basic database is established.
Step S303：Data after fusion are packed into basic database, are uniformly stored in data center.
Step S304：Analyze data.
(such as with the essential information of entity object：The case where entity object is behaved, then the essential information of people includes identity card
Number, passport number, phone number, bank's card number, driver license number, social security number) it is keyword, each entity object is extracted in fusion
Data of the people in social activities.Further, owner's entity object (in 1 year such as nearest) was acquired in some period
All social activities characteristic informations.Using the clustering algorithm of unsupervised learning owner's social activities feature is collected to above-mentioned
Information sort out by similitude, is classified as normal social activities citizen and abnormal social activities 2 classifications of citizen, further, point
It is other to above-mentioned two category set, establish normal social activities citizen behavior mathematical model and abnormal society using model-naive Bayesian
It can movable citizen's behavior mathematical model.
Step S305：Discriminate line is utilized for the suspect participated in case according to the event or case specifically occurred
Property regression model to the characteristic information of suspicious people respectively with normal social activities citizen behavior mathematical model and abnormal social activities
Citizen's behavior mathematical model matches, and predicts the probability that the social activities of suspicious people is suspected of committing a crime, explores the event or case
The information that the solution or detection of part need, help is provided to clear up a criminal case.
Step S306：Data are sounded out the people in a given scope one by one in order to break a criminal case.
On the basis of the social relationships collection of illustrative plates that above-mentioned steps are established, made according to the specific characteristic information that event or case occurs
Sound out the people in a given scope one by one in order to break a criminal case condition for data, the condition sounded out the people in a given scope one by one in order to break a criminal case can be include event occur time, place, the participation of the personage event entity
Characteristic attribute, on social relationships collection of illustrative plates setting sound out the people in a given scope one by one in order to break a criminal case condition, be exactly specifically, arbitrary two entities that social relationships occur it
Between allow setting to sound out the people in a given scope one by one in order to break a criminal case condition, the condition filter that can be sounded out the people in a given scope one by one in order to break a criminal case by setting fall with the event or the incoherent entity information of case,
Then namely denoising carries out denoising, uncorrelated to the event according to the condition of setting to social relationships collection of illustrative plates
The information filtering of people, object fall.It is sounded out the people in a given scope one by one in order to break a criminal case condition by repeatedly setting different data to social relationships collection of illustrative plates repeatedly, exploration obtains
The identical social relationships collection of illustrative plates with the event or case.
Step S307：The entity information for obtaining the event or the relevant people of case, object, can be with ancillary events according to the information
Or the solution and detection of case.
The embodiment of the present invention quickly finds case clue it can be found that association hiding inside data, and automatically structure with
The identical social relationships collection of illustrative plates of the case, makes offender have no place to hide.
It will appreciated by the skilled person that implement the method for the above embodiments be can be with
Relevant hardware is instructed to complete by program, the program can be stored in computer read/write memory medium, should
When being executed, including step S101 is to step S103 for program, either including step S201 to step S205 or including step
S301 to step S307.That is, the present invention can also provide a kind of computer readable storage medium, it is stored thereon with calculating
Machine program, the program at least realize following steps when being executed by processor：According to the information of entity object, structure is based on the reality
The relation map of body object；According to the characteristic information for the event that occurred, determine that being used for handling the data that event has occurred touches
Row's condition；It is sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determines and thing has occurred with described
The related entity object of part.Wherein, the storage medium may include ROM/RAM, magnetic disc, CD, USB flash disk.
Example IV
Fig. 4 is the device block diagram for the exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data that the embodiment of the present invention four provides, such as Fig. 4 institutes
Show, described device includes：
Preprocessing module 401, for the data of the different data sources of entity object to be carried out data scrubbing and data fusion,
Obtain the information of the entity object, wherein the information of the entity object includes that the essential information of entity object and activity are believed
Breath.
Module 402 is built, for the information according to entity object, builds the relation map based on the entity object.Institute
It states structure module 402 to be specifically used for using the essential information of the entity object as keyword, acquisition has with the entity object
Other entity objects of activity relationship, and establish other comprising described in activity relationship centered on the entity object
The relation map of entity object.
Setup module 403 determines and event has occurred described in being used for handling for the characteristic information according to the event that occurred
Data are sounded out the people in a given scope one by one in order to break a criminal case condition.Wherein, the characteristic information that event has occurred includes the time of the generation for the event that occurred, place, relates to
And character features.
Module of sounding out the people in a given scope one by one in order to break a criminal case 404 carries out denoising for sounding out the people in a given scope one by one in order to break a criminal case condition according to the data to constructed relation map,
It determines and the related entity object of event has occurred with described.The module 404 of sounding out the people in a given scope one by one in order to break a criminal case is specifically used for sounding out the people in a given scope one by one in order to break a criminal case item using the data
Part, filter out with the relation map that event has occurred is unrelated, obtain that the related relation map of event has occurred with described, and obtain
Take the corresponding entity object of the relation map.That is, the denoising removes and has been occurred from entity object
The unrelated entity object of event.
Further, it is further secondary locking suspect, the structure module 402 is also for assisting clear up a criminal case
It can be used in the information according to entity object, after building the relation map based on the entity object, according to described
The action message of entity object within a preset period of time carries out similitude classification to the entity object, and for per class entity pair
As establishing the respective crawler behavior model with action message, such as it is divided into the normal society of corresponding normal social activities citizen
The abnormal social activities citizen behavior mathematical model of movable citizen's behavior mathematical model and corresponding abnormal social activities citizen.Into one
Ground is walked, the module 404 of sounding out the people in a given scope one by one in order to break a criminal case can be also used for respective with work to be established per class entity object in the structure module 402
After the crawler behavior model of dynamic information, according to the action message that the entity object that event is related to has occurred, determine described in
Crawler behavior model belonging to action message, and according to definitive result, such as belong to abnormal social activities citizen behavior mathematics
Model predicts the probability that event has occurred described in the entity object participation, to obtain and the advantageous information of clear up a criminal case.
Above-mentioned preprocessing module 401, structure module 402, setup module 403 and module 404 of sounding out the people in a given scope one by one in order to break a criminal case can be applied to political affairs
The intelligence analysis system of the departments such as mansion, public security, criminal investigation, when it is implemented, can be realized by the processor of intelligence analysis system.
Embodiment five
The device for the exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data that embodiment five provides, including：
Processor, for according to the information of entity object, building the relation map based on the entity object, according to having sent out
It makes trouble the characteristic information of part, determines that being used for handling the data that event has occurred sounds out the people in a given scope one by one in order to break a criminal case condition, and sound out the people in a given scope one by one in order to break a criminal case according to the data
Condition carries out denoising to constructed relation map, determines and the related entity object of event has occurred with described；
Memory is coupled with the processor, the program that storage is executed for the processor thereon.
Further, the processor can be also used for the activity letter according to the entity object within a preset period of time
Breath carries out similitude classification to the entity object, and to establish the respective activity with action message per class entity object
Behavior model, in this way, after an event occurs, according to crawler behavior model, can predict that the entity object that the event is related to participates in
To the possibility of the event, advantageous information is provided for quick clear up a criminal case.
In conclusion the embodiment of the present invention has the following technical effects：
The embodiment of the present invention can expand data acquisition channel and range, augmentation data fusion, to greatest extent depth excavation
The internal association of information establishes the stereo data model based on people and case；
The embodiment of the present invention can fully excavate the value that data with existing contains, and smoothness supports police service to study and judge personnel and realize packet
It includes and people is looked for object, case is looked for people, looking for a variety of thinkings of sounding out the people in a given scope one by one in order to break a criminal case of case with case, building corresponding tactics model and keep utilizing, become
Police service studies and judges personnel's development and combines the strong professional tool sounded out the people in a given scope one by one in order to break a criminal case on the net.
This patent and is handled and analysis by mass data collection, is established based on the entity object including people, vehicle
The association of essential information, the automatic sociogram for building object, is then based on the sociogram, user can be according to specific
Affair character, in the social relationships, setting data are sounded out the people in a given scope one by one in order to break a criminal case condition, which are repeated denoising, directly
To obtain with the matched sociogram of the affair character, analyze the valuable data being hidden in mass data, assist
The solution and detection of event (case).
Although describing the invention in detail above, but the invention is not restricted to this, those skilled in the art of the present technique
It can be carry out various modifications with principle according to the present invention.Therefore, all to be changed according to made by the principle of the invention, all it should be understood as
Fall into protection scope of the present invention.
Claims (15)
1. a kind of method for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data, including：
According to the information of entity object, the relation map based on the entity object is built；
According to the characteristic information for the event that occurred, determine that being used for handling the data that event has occurred sounds out the people in a given scope one by one in order to break a criminal case condition；
It is sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determines and event has occurred with described
Related entity object.
2. according to the method described in claim 1, in the information according to entity object, build based on the entity object
Before relation map, including：
Data scrubbing and data fusion are carried out to the data of the different data sources of entity object, obtain the letter of the entity object
Breath, wherein the information of the entity object includes the essential information and action message of entity object.
3. according to the method described in claim 2, the information according to entity object, builds the pass based on the entity object
It is that collection of illustrative plates includes：
Using the essential information of the entity object as keyword, other entities that there is activity relationship with the entity object are obtained
Object, and establish the relational graph for including other entity objects with activity relationship centered on the entity object
Spectrum.
4. according to the method in claim 2 or 3, in the information according to entity object, structure is based on the entity pair
After the relation map of elephant, including：
According to the action message of the entity object within a preset period of time, similitude classification is carried out to the entity object, and
The respective crawler behavior model with action message is established for every class entity object.
5. according to the method described in claim 4, establishing the respective work with action message described for every class entity object
After dynamic behavior model, further comprise：
According to the action message that the entity object that event is related to has occurred, the crawler behavior belonging to the action message is determined
Model, and according to definitive result, predict the probability that event has occurred described in the entity object participation.
6. according to the method described in claim 1, the characteristic information that event has occurred include it is described occurred event when
Between, place, personage, the basis occurred the characteristic information of event, determined that being used for handling the data that event has occurred touches
Row's condition includes：
According at least one of the time that event has occurred, place, personage, determine that the data are sounded out the people in a given scope one by one in order to break a criminal case condition.
7. according to the method described in claim 1, described sound out the people in a given scope one by one in order to break a criminal case condition according to the data, constructed relation map is carried out
Denoising, determines and includes with the related entity object of event that occurred：
Sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, filter out with the relation map that event has occurred is unrelated, obtain having occurred with described
The related relation map of event, and obtain the corresponding entity object of the relation map.
8. a kind of device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data, including：
Module is built, for the information according to entity object, builds the relation map based on the entity object；
Setup module determines that being used for handling the data that event has occurred touches for the characteristic information according to the event that occurred
Row's condition；
It sounds out the people in a given scope one by one in order to break a criminal case module, for sounding out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determine and institute
State the related entity object of the event of generation.
9. device according to claim 8, further comprises：
Preprocessing module is used in the information according to entity object, right before building the relation map based on the entity object
The data of the different data sources of entity object carry out data scrubbing and data fusion, obtain the information of the entity object, wherein
The information of the entity object includes the essential information and action message of entity object.
10. device according to claim 9, the structure module specifically for being with the essential information of the entity object
During keyword obtains the other entity objects for having activity relationship with the entity object, and foundation is with the entity object
The relation map for including other entity objects with activity relationship of the heart.
11. device according to claim 9 or 10, what the structure module was further used for described according to entity object
Information, after building the relation map based on the entity object, according to the activity of the entity object within a preset period of time
Information carries out similitude classification to the entity object, and to establish the respective work with action message per class entity object
Dynamic behavior model.
12. according to the devices described in claim 11, the module of sounding out the people in a given scope one by one in order to break a criminal case is further used for establishing respectively for every class entity object
From the crawler behavior model with action message after, believed according to the activity that entity object that event is related to has occurred
Breath determines the crawler behavior model belonging to the action message, and according to definitive result, predicts described in the entity object participation
The probability of event has occurred.
13. device according to claim 8, the characteristic information that event has occurred include it is described occurred event when
Between, place, personage, the setup module is specifically used for that at least one in time of event, place, personage has occurred according to
It is a, determine that the data are sounded out the people in a given scope one by one in order to break a criminal case condition.
14. device according to claim 8, the module of sounding out the people in a given scope one by one in order to break a criminal case is specifically used for being sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, filters out
It with the relation map that event has occurred is unrelated, obtains that the related relation map of event has occurred with described, and described in obtaining
The corresponding entity object of relation map.
15. a kind of device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data, including：
Memory；And
Processor is coupled to the memory, for the information according to entity object, builds based on the entity object
Relation map determines that being used for handling the data that event has occurred sounds out the people in a given scope one by one in order to break a criminal case condition according to the characteristic information for the event that occurred, and
It is sounded out the people in a given scope one by one in order to break a criminal case condition according to the data, denoising is carried out to constructed relation map, determined and described that event has occurred is related
Entity object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710247747.0A CN108733683A (en) | 2017-04-17 | 2017-04-17 | A kind of method and device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710247747.0A CN108733683A (en) | 2017-04-17 | 2017-04-17 | A kind of method and device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108733683A true CN108733683A (en) | 2018-11-02 |
Family
ID=63925094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710247747.0A Pending CN108733683A (en) | 2017-04-17 | 2017-04-17 | A kind of method and device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108733683A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009022A (en) * | 2019-03-26 | 2019-07-12 | 第四范式（北京）技术有限公司 | Prediction technique, device and the calculating equipment of drug addict's information |
| CN117830060A (en) * | 2024-03-04 | 2024-04-05 | 天津财经大学 | Injury crime law enforcement supervision and auxiliary decision-making system based on knowledge graph |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130262586A1 (en) * | 2012-04-03 | 2013-10-03 | Xtreme Labs, Inc. | Method and System for Filtering Device Events by Social Graph |
| US20140280108A1 (en) * | 2013-03-14 | 2014-09-18 | Jeffrey Dunn | Systems, methods, and apparatuses for implementing an interface to view and explore socially relevant concepts of an entity graph |
| US20150347478A1 (en) * | 2014-06-03 | 2015-12-03 | Xerox Corporation | Systems and methods for context-aware and personalized access to visualizations of road events |
| CN105183869A (en) * | 2015-09-16 | 2015-12-23 | 分众(中国)信息技术有限公司 | Building knowledge mapping database and construction method thereof |
| CN105468605A (en) * | 2014-08-25 | 2016-04-06 | 济南中林信息科技有限公司 | Entity information map generation method and device |
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
-
2017
- 2017-04-17 CN CN201710247747.0A patent/CN108733683A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130262586A1 (en) * | 2012-04-03 | 2013-10-03 | Xtreme Labs, Inc. | Method and System for Filtering Device Events by Social Graph |
| US20140280108A1 (en) * | 2013-03-14 | 2014-09-18 | Jeffrey Dunn | Systems, methods, and apparatuses for implementing an interface to view and explore socially relevant concepts of an entity graph |
| US20150347478A1 (en) * | 2014-06-03 | 2015-12-03 | Xerox Corporation | Systems and methods for context-aware and personalized access to visualizations of road events |
| CN105468605A (en) * | 2014-08-25 | 2016-04-06 | 济南中林信息科技有限公司 | Entity information map generation method and device |
| CN105183869A (en) * | 2015-09-16 | 2015-12-23 | 分众(中国)信息技术有限公司 | Building knowledge mapping database and construction method thereof |
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110009022A (en) * | 2019-03-26 | 2019-07-12 | 第四范式（北京）技术有限公司 | Prediction technique, device and the calculating equipment of drug addict's information |
| CN117830060A (en) * | 2024-03-04 | 2024-04-05 | 天津财经大学 | Injury crime law enforcement supervision and auxiliary decision-making system based on knowledge graph |
| CN117830060B (en) * | 2024-03-04 | 2024-05-28 | 天津财经大学 | Injury crime law enforcement supervision and auxiliary decision-making system based on knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alam et al. | Deep learning benchmarks and datasets for social media image classification for disaster response | |
| Duarte et al. | Vehicle classification in distributed sensor networks | |
| CN111950937A (en) | Key personnel risk assessment method based on fusion space-time trajectory | |
| CN109241223B (en) | Behavior track identification method and system | |
| Sharma | Z-CRIME: A data mining tool for the detection of suspicious criminal activities based on decision tree | |
| Krippner et al. | The person of the category: The pricing of risk and the politics of classification in insurance and credit | |
| CN113641906A (en) | System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data | |
| CN108733683A (en) | A kind of method and device for exploration event clue of being sounded out the people in a given scope one by one in order to break a criminal case based on data | |
| Shi et al. | PR‐NET: progressively‐refined neural network for image manipulation localization | |
| CN115438102A (en) | Space-time data anomaly identification method and device and electronic equipment | |
| Carmichael et al. | Unfooling perturbation-based post hoc explainers | |
| Zhu et al. | Spatiotemporal-textual point processes for crime linkage detection | |
| CN114707685A (en) | Event prediction method and device based on big data modeling analysis | |
| CN112632532B (en) | User abnormal behavior detection method based on deep forest in edge calculation | |
| Zubi et al. | Using data mining techniques to analyze crime patterns in the libyan national crime data | |
| CN110347669A (en) | Risk prevention method based on streaming big data analysis | |
| CN109101568A (en) | Traffic high-risk personnel recognition methods based on XgBoost algorithm | |
| Kaur et al. | Analyzing and displaying of crime hotspots using fuzzy mapping method | |
| Adderley | The use of data mining techniques in operational crime fighting | |
| Niloy et al. | A novel disaster image data-set and characteristics analysis using attention model | |
| CN115510248A (en) | Method for constructing and analyzing person behavior characteristic knowledge graph based on deep learning | |
| CN109063751A (en) | The traffic high-risk personnel recognition methods of decision Tree algorithms is promoted based on gradient | |
| CN114692836A (en) | Block chain assisted personnel negative emotion comprehensive evaluation method and system | |
| CN113469816A (en) | Digital currency identification method, system and storage medium based on multigroup technology | |
| Prasad et al. | Analysis and prediction of crime against woman using machine learning techniques |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181102 |