CN104050239B - Correlation matching analyzing method among multiple objects - Google Patents
Correlation matching analyzing method among multiple objects Download PDFInfo
- Publication number
- CN104050239B CN104050239B CN201410225694.9A CN201410225694A CN104050239B CN 104050239 B CN104050239 B CN 104050239B CN 201410225694 A CN201410225694 A CN 201410225694A CN 104050239 B CN104050239 B CN 104050239B
- Authority
- CN
- China
- Prior art keywords
- record
- matching
- data
- criterion
- relevant matches
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/2433—Query languages
- G06F16/2448—Query languages for particular applications; for extensibility, e.g. user defined types
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a correlation matching analyzing method among multiple objects. The method comprises a first step of arranging given data record sets and recording a matching judging method f according to an application scene definition, and a second step of judging whether n objects meet the requirement of the correlation matching judging method f(m) or not for the given n objects and m matching times according to whether m groups of matching record sets which satisfy the matching judging method f* can be extracted from the data record sets or not, wherein n is greater than or equal to three, and m is greater than or equal to two.
Description
Technical field
The present invention relates to the relevant matches analysis method between computer realm, more particularly to a kind of multiple objects.
Background technology
In the data for the track record such as moving in comprising traffic contact or hotel, the relevant matches analysis side between object
Method repeatedly takes identical car and boat, aircraft order of classes or grades at school to analyzing which object, or passes through, moves in or leave a certain in the close time
Place is this kind of, and to live problem together significant.In fields such as anti-terrorism investigation, strike traffic in drugs, multiple level marketing networks, analysis is needed
Go together or live relation or other related sexual behaviour together to judging whether the colony for reaching certain scale has, that is, study multiple objects
Between relevant matches problem.And current object dependencies the matching analysis technology generally refers to an object, analysis has
Which object is matched more than given number of times with him, or in given scope, is found out all object dependencies matching times and surpassed
Cross set-point combination of two.These analysis modes belong to binary crelation analysis, and whether its core is to judge between two objects
Matching.Because the reference point matched respectively between object is probably different, even if object is matched two-by-two, three can not be proved
There is matching relationship between object above.It means that existing object dependencies Match Analysis can not meet instead
Probably the needs of behavioural analysis judgement are lived in the field such as investigation, strike traffic in drugs, multiple level marketing network together to many people.Accordingly, it would be desirable to be based on
Multiple objects are studying relevant matches problem.And the relevant matches between multiple objects are studied, face matching and judge regulation
Then, the diversity of combination is matched between matching times and triple uncertain and objects of number of objects, and amount of calculation is with data
Scale exponentially increases, and needs those skilled in the art to pay creative work, could solve corresponding technical problem.
The content of the invention
It is contemplated that at least solving technical problem present in prior art, it is a kind of multiple right especially innovatively to propose
Relevant matches analysis method as between.
In order to realize the above-mentioned purpose of the present invention, the invention provides the relevant matches analysis between a kind of multiple objects
Method, it is critical only that：Comprise the steps：
Step 1, the data record set to giving is arranged, and matches criterion f according to application scenarios definition record；
Can step 2, to given n object and matching times m, according to extract from data record set matching be met
Criterion f*M groups matching record set, judge whether the n object meets relevant matches criterion f(m)Requirement, wherein
Described n >=3, m >=2.
Above-mentioned technical proposal has the beneficial effect that：After the present invention is by accurate screening matching work, can position and reach
To the data of threshold range, corresponding object can be embodied accurately, without the need for the i.e. achievable accurate match of manual operation.
Calculated using alternative manner during Data Matching, can preferably screen object data carries out correlation
Property matching.
Relevant matches analysis method between described multiple objects, it is preferred that include：In the data record set
Every data be recorded as one of object p and possess, R (p) represents the set of the data record that object p possesses；
Record matching criterion f is the Boolean function based on the binary crelation of data record, if two records are respectively ri
And rj, then f (ri,rj)=true represents riAnd rjMatching, f (ri,rj)=false represents riAnd rjMismatch, wherein to record
Definition with criterion f should meet reflexivity and symmetry, i.e., to same data record r, f (r, r) ≡ true, to any two
Data records riAnd rj, there is f (ri,rj)≡f(rj,ri)；
Matching criterion f*It is the Boolean function based on the n-tuple relation of data record, its input is a record set,
It is required that the record of the pieces of data in record set adheres to different objects separately, to record set R, if any two records r in RiAnd rj,
There are f (ri,rj)=true, then claim R to meet matching criterion f*, it is designated as f*(R)=true, and R is called one group of matching record
Collection；
Relevant matches criterion f(m)For the Boolean function of object-based n-tuple relation, its input is an object
Set, if object set is P, if different satisfaction matching criterion f of m groups can be found*Record set Ri, and RiIn data note
Record is belonging respectively to each object in P, i.e.,
The object in P is then claimed to meet relevant matches criterion f(m), it is designated as f(m)(P)=true.
Above-mentioned technical proposal has the beneficial effect that：Being calculated by object matching method needs the data of matching, and data are entered
Row computing, can more accurately carry out object matching.
Relevant matches analysis method between described multiple objects, it is preferred that the step 2 comprises the steps：
Step 2-1, definition includes the set P for specifying n object, and each object in P is filtered out from data record set
All data records, go to step 2-2；
Step 2-2, if the quantity of data record that certain object possesses judges that this n object is unsatisfactory for phase less than m
Closing property matching criterion f(m), terminate, otherwise go to step 2-3；
Step 2-3, if also meeting transitivity based on matching criterion f of data record, i.e., for any 3 data is recorded
r1、r2、r3, have
2-4 is then gone to step, 2-7 is otherwise gone to step；
Step 2-4, by the data record for filtering out some subsets are divided into, it is desirable to each subset Ri, meet f*(Ri)=
True, goes to step 2-5；
Step 2-5, initially puts and is counted as 0, to each subset RiIf to any object p in P, having
Then count is incremented, goes to step 2-6；
Whether step 2-6, m is not less than according to tale, judges whether the n object meets relevant matches criterion
f(m)Requirement, terminate；
Step 2-7, selects an object p from P0, create temporary object set P'={ p0, it is p0Possess per bar number
Matching record set R={ r } is respectively created according to record r, 2-8 is gone to step；
Step 2-8, if P'=P, judges that the n object dependencies match criterion f(m)Requirement, otherwise from P
Another object p of reselection, goes to step 2-9；
Step 2-9, checks one by one each matching record set R, selects data record r to meet f from R (p)*(R∪{r})
=true, makes R=R ∪ { r }, if can not find such data record from R (p), deletes R, treats all matching record sets
Inspection goes to step 2-10 after finishing；
Step 2-10, if the quantity of current matching record set is not less than m, makes P'=P' ∪ { p }, goes to step 2-8, no
Then judge that this n object does not meet relevant matches requirement, terminate.
Above-mentioned technical proposal has the beneficial effect that：Matched after work by accurate screening, can position and reach threshold value
The data of scope, corresponding object can be embodied accurately, and without the need for manual operation accurate match is capable of achieving.With many
Relevant matches analysis method between individual object, is conducive to action trail closely related between Finding Object.
Relevant matches analysis method between described multiple objects, it is preferred that also include：
N object is sorted from less to more by the quantity comprising record, record less object be first added to it is interim right
As in set P'.
Above-mentioned technical proposal has the beneficial effect that：With the relevant matches analysis method between multiple objects, be conducive to
Closely related action trail between Finding Object.
Relevant matches analysis method between described multiple objects, it is preferred that also include：
If data record matching criterion f is judged based on temporally adjacent principle, i.e., per data, record r has
Time attribute τ (r) and time-domain δ (r) is allowed, for meeting f (ri,rjThe two datas record r of)=trueiAnd rj, only work as τ
(ri)∈δ(rj) and τ (rj)∈δ(ri), thus, it is per group and matches the permission time-domain that record set R safeguards wherein all records
Common factor δ*(R), f is judged to arbitrarily record r*(R ∪ { r })=true first checks whether there is τ (r) ∈ δ when whether setting up*(R), then
R can be avoided to compare one by one with the record in R.
Above-mentioned technical proposal has the beneficial effect that：With the relevant matches analysis method between multiple objects, be conducive to
Closely related action trail between Finding Object.
Relevant matches analysis method between described multiple objects, it is preferred that also include：
In multithreading or distributed environment, object to be judged is divided into some subsets and is judged respectively, after remerging
Judge whether all objects meet relevant matches requirement.
In sum, as a result of above-mentioned technical proposal, the invention has the beneficial effects as follows：
After the present invention is by accurate screening matching work, the data for reaching threshold range can be positioned, it is corresponding right
As can accurately embody, without the need for manual operation accurate match is capable of achieving.
Calculated using alternative manner during Data Matching, can preferably screen object data carries out correlation
Property matching.
The present invention is applied to railway police department and the network supervision Ministry of Public Security multiple object dependencies the matching analysis
Door.
With the relevant matches analysis method between multiple objects, be conducive to behavior closely related between Finding Object
Track.
The additional aspect and advantage of the present invention will be set forth in part in the description, and partly will become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Description of the drawings
The above-mentioned and/or additional aspect and advantage of the present invention will become from the description with reference to accompanying drawings below to embodiment
It is substantially and easy to understand, wherein：
Fig. 1 is the schematic diagram of the relevant matches analysis method between multiple objects of the invention；
Fig. 2 is the schematic diagram of the relevant matches analysis method specific embodiment between multiple objects of the invention.
Specific embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from start to finish
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not considered as limiting the invention.
In describing the invention, it is to be understood that term " longitudinal direction ", " horizontal ", " on ", D score, "front", "rear",
The orientation or position relationship of the instruction such as "left", "right", " vertical ", " level ", " top ", " bottom " " interior ", " outward " is based on accompanying drawing institute
The orientation for showing or position relationship, are for only for ease of the description present invention and simplify description, rather than indicate or imply the dress of indication
Put or element must have specific orientation, with specific azimuth configuration and operation, therefore it is not intended that to the present invention limit
System.
In describing the invention, unless otherwise prescribed and limit, it should be noted that term " installation ", " connected ",
" connection " should be interpreted broadly, for example, it may be mechanically connected or electrical connection, or the connection of two element internals, can
Being to be joined directly together, it is also possible to be indirectly connected to by intermediary, for the ordinary skill in the art, can basis
Concrete condition understands the concrete meaning of above-mentioned term.
As shown in Figure 1, 2, the invention provides the relevant matches analysis method between a kind of multiple objects, its key exists
In：Comprise the steps：
Step 1, the data record set to giving is arranged, and matches criterion f according to application scenarios definition record；
Can step 2, to given n object and matching times m, according to extract from data record set matching be met
Criterion f*M groups matching record set, judge whether the n object meets relevant matches criterion f(m)Requirement, wherein
Described n >=3, m >=2.
Above-mentioned technical proposal has the beneficial effect that：After the present invention is by accurate screening matching work, can position and reach
To the data of threshold range, corresponding object can be embodied accurately, without the need for the i.e. achievable accurate match of manual operation.
Calculated using alternative manner during Data Matching, can preferably screen object data carries out correlation
Property matching.
Relevant matches analysis method between described multiple objects, it is preferred that include：In the data record set
Every data be recorded as one of object p and possess, R (p) represents the set of the data record that object p possesses；
Record matching criterion f is the Boolean function based on the binary crelation of data record, if two record difference
For riAnd rj, then f (ri,rj)=true represents riAnd rjMatching, f (ri,rj)=false represents riAnd rjMismatch, wherein to note
The definition of record matching criterion f should meet reflexivity and symmetry, i.e., to same data record r, f (r, r) ≡ true, to appointing
Two datas of anticipating record riAnd rj, there is f (ri,rj)≡f(rj,ri)；
Matching criterion f*It is the Boolean function based on the n-tuple relation of data record, its input is a record set,
It is required that the record of the pieces of data in record set adheres to different objects separately, to record set R, if any two records r in RiAnd rj,
There are f (ri,rj)=true, then claim R to meet matching criterion f*, it is designated as f*(R)=true, and R is called one group of matching record
Collection；
Relevant matches criterion f(m)For the Boolean function of object-based n-tuple relation, its input is an object
Set, if object set is P, if different satisfaction matching criterion f of m groups can be found*Record set Ri, and RiIn data note
Record is belonging respectively to each object in P, i.e.,
The object in P is then claimed to meet relevant matches criterion f(m), it is designated as f(m)(P)=true.
Above-mentioned technical proposal has the beneficial effect that：Being calculated by object matching method needs the data of matching, and data are entered
Row computing, can more accurately carry out object matching.
Relevant matches analysis method between described multiple objects, it is preferred that the step 2 comprises the steps：
Step 2-1, definition includes the set P for specifying n object, and each object in P is filtered out from data record set
All data records, go to step 2-2；
Step 2-2, if the quantity of data record that certain object possesses judges that this n object is unsatisfactory for phase less than m
Closing property matching criterion f(m), terminate, otherwise go to step 2-3；
Step 2-3, if also meeting transitivity based on matching criterion f of data record, i.e., for any 3 data is recorded
r1、r2、r3, have
2-4 is then gone to step, 2-7 is otherwise gone to step；
Step 2-4, by the data record for filtering out some subsets are divided into, it is desirable to each subset Ri, meet f*(Ri)=
True, goes to step 2-5；
Step 2-5, initially puts and is counted as 0, to each subset RiIf to any object p in P, havingThen count is incremented, goes to step 2-6；
Whether step 2-6, m is not less than according to tale, judges whether the n object meets relevant matches criterion
f(m)Requirement, terminate；
Step 2-7, selects an object p from P0, create temporary object set P'={ p0, it is p0Possess per bar number
Matching record set R={ r } is respectively created according to record r, 2-8 is gone to step；
Step 2-8, if P'=P, judges that the n object dependencies match criterion f(m)Requirement, otherwise from P
Another object p of reselection, goes to step 2-9；
Step 2-9, checks one by one each matching record set R, selects data record r to meet f from R (p)*(R∪{r})
=true, makes R=R ∪ { r }, if can not find such data record from R (p), deletes R, treats all matching record sets
Inspection goes to step 2-10 after finishing；
Step 2-10, if the quantity of current matching record set is not less than m, makes P'=P' ∪ { p }, goes to step 2-8, no
Then judge that this n object does not meet relevant matches requirement, terminate.
Above-mentioned technical proposal has the beneficial effect that：Matched after work by accurate screening, can position and reach threshold value
The data of scope, corresponding object can be embodied accurately, and without the need for manual operation accurate match is capable of achieving.With many
Relevant matches analysis method between individual object, is conducive to action trail closely related between Finding Object.
Relevant matches analysis method between described multiple objects, it is preferred that also include：
N object is sorted from less to more by the quantity comprising record, record less object be first added to it is interim right
As in set P'.
Above-mentioned technical proposal has the beneficial effect that：With the relevant matches analysis method between multiple objects, be conducive to
Closely related action trail between Finding Object.
Relevant matches analysis method between described multiple objects, it is preferred that also include：
If data record matching criterion f is judged based on temporally adjacent principle, i.e., per data, record r has
Time attribute τ (r) and time-domain δ (r) is allowed, for meeting f (ri,rjThe two datas record r of)=trueiAnd rj, only work as τ
(ri)∈δ(rj) and τ (rj)∈δ(ri), thus, it is per group and matches the permission time-domain that record set R safeguards wherein all records
Common factor δ*(R), f is judged to arbitrarily record r*(R ∪ { r })=true first checks whether there is τ (r) ∈ δ when whether setting up*(R), then
R can be avoided to compare one by one with the record in R.
Above-mentioned technical proposal has the beneficial effect that：With the relevant matches analysis method between multiple objects, be conducive to
Closely related action trail between Finding Object.
Relevant matches analysis method between described multiple objects, it is preferred that also include：
In multithreading or distributed environment, object to be judged is divided into some subsets and is judged respectively, after remerging
Judge whether all objects meet relevant matches requirement.
Include before the step 1：
When carrying out object authentication between multiple objects, by object authentication information transfer to cloud server, the high in the clouds
The identity information of the plurality of object is transferred to analysing terminal by server, and behavior is carried out to the plurality of object by analysing terminal
Matching.
Also include after step 2：
The data transfer in the matching record set after matching be will determine that to management terminal, when the plurality of object matching degree
After reaching respective threshold, administrative staff are reminded.
The invention is with regard to the relevant matches analysis method between multiple objects.If a group objects is repeatedly together
Identical car and boat, aircraft order of classes or grades at school are taken, or passed through, moved in or leave a certain place (order of classes or grades at school, the place of each time in the close time
Can be with difference), then we claim the group objects to have record relevant matches relation.Judge and find the correlation between multiple objects
The process of property matching relationship is the relevant matches analysis method between multiple objects.
The condition of the relevant matches analysis method between multiple objects is as follows：
1. the set of all objects investigated and the set of all object whereabouts, every track record comprising the time,
The key elements such as point (or order of classes or grades at school).Such as in the data of public online place the matching analysis, per bar record comprising online people's identity information,
The information such as upper off line time, place Internet bar.
2. whether any two records meet the judgment rule of object dependencies matching.Such as in flight colleague's analysis, sentence
It is identical that disconnected rule is usually flight.And for example in railway booking colleague's analysis, one of which judgment rule is the train class taken
Secondary identical and booking time phase difference is within 30 seconds.
3. the minimum number (being set to m) of relevant matches is recorded, generally should be more than 2 times.
The data acquisition of plurality of object behavior track calls at any time this in object database by cloud server
A little object datas, and the collection of object data action trail passes through, flight data system connection cloud server, the high in the clouds clothes
Business device connecting object database, rds data system connection cloud server, the cloud server connecting object database, visitor
Fortune data system connection cloud server, the cloud server connecting object database, highway data system connection cloud
End server, the cloud server connecting object database, bank data system connection cloud server, the cloud service
Device connecting object database.
The action trail data message of above-mentioned data system acquisition target, in being transferred to object database, object data
According to the demand of user, selecting needs the object of screening matching in storehouse, carries out the relevant matches analysis between multiple objects.
The data message of above-mentioned object, the information that can be scanned by identity card, or finger print information, pupil information is carried out
Object confirms.
The determination methods of the relevant matches analysis method between multiple objects are as follows：
1. a pair given group objects, if we find out one-to-one one group of track record, and appoints during these are recorded
Meaning two all meets the judgment rule of above-mentioned record relevant matches, then we claim this group of people to have a relevant matches, claims
This group of track record is one group of matching record set of this group objects.
2. if we can find out the diverse matching record set correspondence track record of m groups for the group objects, and per group
In any two judgment rules for all meeting above-mentioned record relevant matches, then we claim the group objects have multiple objects it
Between relevant matches.
Verify whether multiple objects (being set to n) have the determination methods of relevant matches relation：
1. directly judged according to rule, be that this n people finds out successively m groups matching record set.
2. judged in an iterative manner.
First judge whether that at least m matching record set can be found for 2 objects therein, then see again and add another
After object, if remain to find at least m matching record set for these objects.Like this, object is added to one by one in set.
If can find at least m matching record set all the time, we judge that this n object has the correlation between multiple objects
With analysis.
Add from the point of view of us now after an object and how to look for matching record set.If front k-1 object has m1 correlation
Matching, that is, have m1 groups matching record set (m1 is not less than m).Now we are the relevant matches relation that judges this k object
No establishment.Match record set to per group of front k-1 object, we from the action trail record of latter object, search with
The record for all meeting is recorded per bar in this group matching record set.K is more than or equal to 1.
If found, this record is matched other records in record set with the group and is combined, as this k
One group of matching record set of object.If at least m groups matching record set, the relevant matches relation of this k object can be found
Set up.
Whole deterministic process refers to Fig. 2.
S1, is input into n object to be determined, and the n is more than or equal to 3；
S2, obtains the action trail record of each object；
S3, arranges k=1, and the k is recorded for the action trail of object；
S4, is that each object records matching record set of the establishment comprising the record per bar, if matching record set quantity is m1；
3. this n object is divided into some subsets, first checks whether each subset meets the correlation between multiple objects
The relation of the matching analysis, then reexamines the relevant matches relation between subset.This method facilitates implementation parallel computation.
According to the difference of use environment, determination methods can consider to be optimized from the following aspects：
1. track record can temporally or order of classes or grades at school is ranked up, it means that can specify time range in moment positioning
Or the track record of specified order of classes or grades at school.
If based on order of classes or grades at school, corresponding record number is less than x bars for the judgement of the relevant matches relation between 2. recording
Order of classes or grades at school can neglect.Similarly, if the judgement of the relevant matches relation between record is based on temporally adjacent principle
, then number being recorded in the range of adjacent time can be ignored less than the record of x bars.The x is more than or equal to 2.
3. under normal circumstances, the more object of action trail is higher with the possibility that other object dependencies are matched.Thus,
If judging whether n object has relevant matches relation, this n object is arranged can from less to more by the quantity of track record
Whether sequence, first judge to record between less object with relevant matches relation.
If the judgement of the relevant matches relation between 4. recording is equal based on attribute (such as date, order of classes or grades at school, place)
Principle, then these attributes of all records are identicals in one group of matching record set.We can be per group of matching record set
Record these attributes.So, newly one object of addition when, we only need to these attributes in the track record by him and match
These attributes of record set are compared just can be with without comparing one by one with the record in matching record set.
If the judgement of the relevant matches relation between 5. recording is recorded all based on temporally adjacent principle per bar
One permission time range of correspondence, record within the range could therewith meet relevant matches relation.Thus, we can be with
It is that per group of matching record set safeguards a permission time range, it is the friendship of the permission time range of all records in the record set
Collection.So, newly one object of addition when, we only need to check the track record of the object whether belong to front k-1 objects
Permission time range with record set, without comparing one by one with the record in matching record set.
The invention has the beneficial effects as follows：
After the present invention is by accurate screening matching work, the data for reaching threshold range can be positioned, it is corresponding right
As can accurately embody, without the need for manual operation accurate match is capable of achieving.
Calculated using alternative manner during Data Matching, can preferably screen object data carries out correlation
Property matching.
The present invention is applied to security department multiple object dependencies the matching analysis.
With the relevant matches analysis method between multiple objects, be conducive to behavior closely related between Finding Object
Track, so as to accurately judge relevant matches data.
Now it is listed below embodiment：
In track record, per bar, record includes 3 fields：Record number, date and record party's name.Wherein Zhao
Certain, money, Sun, Lee have 6,6,5,4 records respectively.If the minimum number that requirement is lived together is 3, between two records
Criterion with the relation of living together has two kinds of situations：1st, the date is identical；2nd, date difference is within one day.By the present invention
The action trail of corresponding object can be clearly matched, is easy to the matching degree of behavior between Finding Object.
| Record number | Date | Name |
| 1 | 2014-03-02 | Zhao |
| 2 | 2014-03-02 | Money |
| 3 | 2014-03-02 | Sun |
| 4 | 2014-03-26 | Zhao |
| 5 | 2014-03-26 | Money |
| 6 | 2014-03-26 | Lee |
| 7 | 2014-04-16 | Zhao |
| 8 | 2014-04-16 | Sun |
| 9 | 2014-04-16 | Lee |
| 10 | 2014-05-01 | Zhao |
| 11 | 2014-05-01 | Money |
| 12 | 2014-05-01 | Sun |
| 13 | 2014-05-21 | Money |
| 14 | 2014-05-21 | Sun |
| 15 | 2014-05-21 | Lee |
| 16 | 2014-06-18 | Zhao |
| 17 | 2014-06-18 | Money |
| 18 | 2014-06-18 | Lee |
| 19 | 2014-06-30 | Zhao |
| 20 | 2014-06-30 | Money |
| 21 | 2014-06-30 | Sun |
1. when using the date it is identical as record between matching judgment standard when
Matching decision rule between now recording meets transferability, can be analyzed by the way of dividing subset.
For Zhao, money, the people of Sun 3, we are divided into by date 6 subsets their record, wherein can look for
Go out 3 groups of matching record sets, first group containing record 1,2,3, second group containing record 10,11,12, the 3rd group containing record 19,20,21,
Therefore there are many people to live relation together for Zhao, Sun, the people of money 3.
For Zhao, money, Lee, we are divided into by date 6 subsets their record, wherein being only capable of finding out 2
Group matching record set, first group contains record 4,5,6, and second group is not gone together containing record 16,17,18, therefore three people with many people
Relation.
2. when within being differed one day using the date as the matching judgment standard between record
Matching decision rule between now recording is unsatisfactory for transferability, can adopt being analyzed for iteration.
Judge whether Zhao, money, the people of Sun 3 there are many people to live relation together.We create 6 from the beginning of Zhao, first
Individual matching record set A1, B1, C1, D1, E1, F1, respectively comprising record 1,4,7,10,16,19.We are again added to money
Come.To matching record set A1, B1, D1, E1, F1, we find and meet the record 2,5,11,17,20 for living rule together, thus
5 matchings record set A2, B2, D2, E2, F2 are created for Zhao and money, wherein A2 includes record 4,5 comprising record 1,2, B2,
D2 is comprising record 10,11, E2 comprising record 16,17, F2 comprising record 19,20.Then we are added Sun.To matching
Record set A2, D2, F2, we find it is corresponding record 3,12,21, be thus Zhao, money, Sun create 3 matching record
Collection A3, D3, F3, wherein A3 is comprising record 1,2,3, D3 comprising record 10,11,12, F3 comprising record 19,20,21.Thus may be used
Know, there are many people to live relation together for Zhao, Sun, the people of money 3.
It will be understood by those skilled in the art that above-mentioned method and apparatus can using computer executable instructions and/or
It is included in processor control routine to realize, such as in the such as mounting medium of disk, CD or DVD-ROM, such as read-only deposits
Provide on the programmable memory of reservoir (firmware) or the data medium of such as optics or electrical signal carrier such
Code.The security setting and check device of the present embodiment and its unit, component can by such as super large-scale integration or
The semiconductor of gate array, logic chip, transistor etc. or such as field programmable gate array, programmable logic device
Deng programmable hardware device hardware circuit realize, it is also possible to by various types of computing devices software realize,
Can being implemented in combination with by above-mentioned hardware circuit and software.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means to combine specific features, structure, material or spy that the embodiment or example are described
Point is contained at least one embodiment of the present invention or example.In this manual, to the schematic representation of above-mentioned term not
Necessarily refer to identical embodiment or example.And, the specific features of description, structure, material or feature can be any
One or more embodiments or example in combine in an appropriate manner.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not
These embodiments can be carried out with various changes, modification, replacement and modification in the case of the principle and objective that depart from the present invention, this
The scope of invention is limited by claim and its equivalent.
Claims (5)
1. the relevant matches analysis method between a kind of multiple objects, it is characterised in that：Comprise the steps：
Step 1, the data record set to giving is arranged, and matches criterion f according to application scenarios definition record；
Can step 2, to given n object and matching times m, according to extract from data record set matching judgement be met
Method f*M groups matching record set, judge whether the n object meets relevant matches criterion f(m)Requirement, wherein described
N >=3, m >=2；
Every data in the data record set is recorded as one of object p and possesses, and R (p) represents the number that object p possesses
According to the set of record；
Record matching criterion f is the Boolean function based on the binary crelation of data record, if two records are respectively riAnd rj,
Then f (ri,rj)=true represents riAnd rjMatching, f (ri,rj)=false represents riAnd rjMismatch, wherein sentencing to record matching
Determining the definition of method f should meet reflexivity and symmetry, i.e., to same data record r, f (r, r) ≡ true, to any two numbers
According to record riAnd rj, there is f (ri,rj)≡f(rj,ri)；
Matching criterion f*It is the Boolean function based on the n-tuple relation of data record, its input is a record set, it is desirable to remembered
The pieces of data record that record is concentrated adheres to different objects separately, to record set R, if any two records r in RiAnd rj, there is f
(ri,rj)=true, then claim R to meet matching criterion f*, it is designated as f*(R)=true, and R is called one group of matching record set；
Relevant matches criterion f(m)For the Boolean function of object-based n-tuple relation, its input is an object set,
If object set is P, if different satisfaction matching criterion f of m groups can be found*Record set Ri, and RiIn data record difference
Belong to each object in P, i.e.,
The object in P is then claimed to meet relevant matches criterion f(m), it is designated as f(m)(P)=true.
2. the relevant matches analysis method between multiple objects according to claim 1, it is characterised in that the step
2 comprise the steps：
Step 2-1, definition includes the set P for specifying n object, and the institute of each object in P is filtered out from data record set
There is data record, go to step 2-2；
Step 2-2, if the quantity of data record that certain object possesses judges that this n object is unsatisfactory for correlation less than m
Matching criterion f(m), terminate, otherwise go to step 2-3；
Step 2-3, if also meeting transitivity based on matching criterion f of data record, i.e., for any 3 data records r1、
r2、r3, have
2-4 is then gone to step, 2-7 is otherwise gone to step；
Step 2-4, by the data record for filtering out some subsets are divided into, it is desirable to each subset Ri, meet f*(Ri)=
True, goes to step 2-5；
Step 2-5, initially puts and is counted as 0, to each subset RiIf to any object p in P, having
Then count is incremented, goes to step 2-6；
Whether step 2-6, m is not less than according to tale, judges whether the n object meets relevant matches criterion f(m)
Requirement, terminate；
Step 2-7, selects an object p from P0, create temporary object set P'={ p0, it is p0The every data note for possessing
Record r is respectively created matching record set R={ r }, goes to step 2-8；
Step 2-8, if P'=P, judges that the n object dependencies match criterion f(m)Requirement, otherwise select again from P
Another object p is selected, 2-9 is gone to step；
Step 2-9, checks one by one each matching record set R, selects data record r to meet f from R (p)*(R ∪ { r })=
True, makes R=R ∪ { r }, if can not find such data record from R (p), deletes R, treats all matching record set inspections
Look into after finishing and go to step 2-10；
Step 2-10, if the quantity of current matching record set is not less than m, makes P'=P' ∪ { p }, goes to step 2-8, otherwise sentences
Disconnected this n object does not meet relevant matches requirement, terminates.
3. the relevant matches analysis method between multiple objects according to claim 2, it is characterised in that also include：
N object is sorted from less to more by the quantity comprising record, less object is recorded and is first added to temporary object collection
In closing P'.
4. the relevant matches analysis method between multiple objects according to claim 2, it is characterised in that also include：
If data record matching criterion f is judged based on temporally adjacent principle, i.e., per data, record r has the time
Attribute τ (r) and time-domain δ (r) is allowed, for meeting f (ri,rjThe two datas record r of)=trueiAnd rj, only as τ (ri)∈
δ(rj) and τ (rj)∈δ(ri), it is the common factor δ of the permission time-domain that per group of matching record set R safeguards wherein all records thus*
(R), f is judged to arbitrarily record r*(R ∪ { r })=true first checks whether there is τ (r) ∈ δ when whether setting up*(R), then r can be avoided
Compare one by one with the record in R.
5. the relevant matches analysis method between multiple objects according to claim 2, it is characterised in that also include：
In multithreading or distributed environment, object to be judged is divided into some subsets and is judged respectively, judged after remerging
Whether all objects meet relevant matches requirement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410225694.9A CN104050239B (en) | 2014-05-27 | 2014-05-27 | Correlation matching analyzing method among multiple objects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410225694.9A CN104050239B (en) | 2014-05-27 | 2014-05-27 | Correlation matching analyzing method among multiple objects |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104050239A CN104050239A (en) | 2014-09-17 |
| CN104050239B true CN104050239B (en) | 2017-04-12 |
Family
ID=51503072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410225694.9A Active CN104050239B (en) | 2014-05-27 | 2014-05-27 | Correlation matching analyzing method among multiple objects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104050239B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107871111B (en) * | 2016-09-28 | 2021-11-26 | 苏宁易购集团股份有限公司 | Behavior analysis method and system |
| CN108197140A (en) * | 2017-11-24 | 2018-06-22 | 中国电子科技集团公司电子科学研究院 | Method for digging, server and the computer readable storage medium of space-time co-occurrence |
| CN108196885B (en) * | 2017-11-27 | 2021-07-27 | 腾讯科技（上海）有限公司 | Method and device for determining object group, storage medium, and electronic device |
| CN111723616B (en) * | 2019-03-20 | 2023-06-02 | 杭州海康威视系统技术有限公司 | Personnel correlation measurement method and device |
| CN110636258B (en) * | 2019-09-09 | 2021-03-02 | 四川东方网力科技有限公司 | Method, device, equipment and storage medium for analyzing peer personnel |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177066A (en) * | 2011-12-20 | 2013-06-26 | Sap门户以色列有限公司 | Analyzing and representing interpersonal relations |
| CN103428164A (en) * | 2012-05-15 | 2013-12-04 | 腾讯科技（深圳）有限公司 | User social network relationship division method and system |
| CN103778151A (en) * | 2012-10-23 | 2014-05-07 | 阿里巴巴集团控股有限公司 | Method and device for identifying characteristic group and search method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080201314A1 (en) * | 2007-02-20 | 2008-08-21 | John Richard Smith | Method and apparatus for using multiple channels of disseminated data content in responding to information requests |
-
2014
- 2014-05-27 CN CN201410225694.9A patent/CN104050239B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177066A (en) * | 2011-12-20 | 2013-06-26 | Sap门户以色列有限公司 | Analyzing and representing interpersonal relations |
| CN103428164A (en) * | 2012-05-15 | 2013-12-04 | 腾讯科技（深圳）有限公司 | User social network relationship division method and system |
| CN103778151A (en) * | 2012-10-23 | 2014-05-07 | 阿里巴巴集团控股有限公司 | Method and device for identifying characteristic group and search method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104050239A (en) | 2014-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bordier et al. | Graph analysis and modularity of brain functional connectivity networks: searching for the optimal threshold | |
| Sidone et al. | Scholarly publication and collaboration in B razil: The role of geography | |
| CN104050239B (en) | Correlation matching analyzing method among multiple objects | |
| Faroqi et al. | Inferring trip purpose by clustering sequences of smart card records | |
| Oliveira et al. | Automated monitoring of construction sites of electric power substations using deep learning | |
| Hong et al. | Assigning passenger flows on a metro network based on automatic fare collection data and timetable | |
| Xie et al. | Testing the proportionality condition with taxi trajectory data | |
| Sherly et al. | BOAT adaptive credit card fraud detection system | |
| Blasi et al. | Analysis of students' misconducts in higher education using decision tree and ann algorithms | |
| Wang et al. | A multi-tier inspection queueing system with finite capacity for differentiated border control measures | |
| Farkas et al. | Voluntary Associations' Impact on the Composition of Active Members' Social Networks: Not an Either/or Matter | |
| Zhang et al. | Dynamic time warp-based clustering: Application of machine learning algorithms to simulation input modelling | |
| Vitello et al. | Transitcrowd: Estimating subway stations demand with mobile crowdsensing data | |
| Yanushkevich et al. | Understanding and taxonomy of uncertainty in modeling, simulation, and risk profiling for border control automation | |
| CN108932530A (en) | The construction method and device of label system | |
| Hu et al. | Individual Travel Knowledge Graph‐Based Public Transport Commuter Identification: A Mixed Data Learning Approach | |
| Zheng et al. | Air transportation direct share analysis and forecast | |
| van Koppen et al. | When to refrain from using likelihood surface methods for geographic offender profiling: An ex ante test of assumptions | |
| CN109472694A (en) | A kind of suspicious trading activity discovery system | |
| Srivastava | Dempster-Shafer theory of belief functions: a language for managing uncertainties in the real-world problems | |
| Fang et al. | Machine learning in facilities & asset management | |
| Garg et al. | Analysis of T‐Spherical Fuzzy Matrix and Their Application in Multiattribute Decision‐Making Problems | |
| Arteaga | The cost of bad parents: Evidence from the effects of incarceration on children’s education | |
| Ai et al. | An overview of prospect theory in travel choice behavior under risk | |
| Ding et al. | OD matching of metro IC card data based on analysis function |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Correlation matching analyzing method among multiple objects Effective date of registration: 20180116 Granted publication date: 20170412 Pledgee: Chongqing Fumin bank Limited by Share Ltd Pledgor: Chongqing Aisi Wang'an Information Technology Co., Ltd.|Zhiduo Information Development Co Ltd, Chongqing Registration number: 2018990000046 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20190319 Granted publication date: 20170412 Pledgee: Chongqing Fumin bank Limited by Share Ltd Pledgor: Chongqing Aisi Wang'an Information Technology Co., Ltd.|Zhiduo Information Development Co Ltd, Chongqing Registration number: 2018990000046 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Correlation matching analyzing method among multiple objects Effective date of registration: 20200110 Granted publication date: 20170412 Pledgee: Chongqing Fumin bank Limited by Share Ltd Pledgor: Chongqing AISI Network Information Technology Co., Ltd.|Chongqing Zhiduo Information Development Co., Ltd Registration number: Y2020990000035 |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 401121 left, 7th floor, building A5, development area, No. 30, Qingfeng North Road, new North District, Chongqing Patentee after: Aisi Technology (Chongqing) Group Co.,Ltd. Patentee after: CHONGQING ZHIDUOXIN INFORMATION DEVELOPMENT Co.,Ltd. Address before: 401121 left, 7th floor, building A5, development area, No. 30, Qingfeng North Road, new North District, Chongqing Patentee before: CHONGQING ISOFT TECHNOLOGY Co.,Ltd. Patentee before: CHONGQING ZHIDUOXIN INFORMATION DEVELOPMENT Co.,Ltd. |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20210324 Granted publication date: 20170412 Pledgee: CHONGQING FUMIN BANK Co.,Ltd. Pledgor: CHONGQING ISOFT TECHNOLOGY Co.,Ltd.|CHONGQING ZHIDUOXIN INFORMATION DEVELOPMENT Co.,Ltd. Registration number: Y2020990000035 |