CN109902934A - City personnel's compartmentalization based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities management method and system - Google Patents
City personnel's compartmentalization based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities management method and system Download PDFInfo
- Publication number
- CN109902934A CN109902934A CN201910087940.1A CN201910087940A CN109902934A CN 109902934 A CN109902934 A CN 109902934A CN 201910087940 A CN201910087940 A CN 201910087940A CN 109902934 A CN109902934 A CN 109902934A
- Authority
- CN
- China
- Prior art keywords
- region
- item
- frequent
- ensure effective
- effective monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
It deploys to ensure effective monitoring and control of illegal activities management method the invention discloses city personnel's compartmentalization based on multi-source big data, this method comprises: obtaining the population development data of multiple regions from multi-source data;According to population development data, the region that the same target and each same target for determining to come across more than one region occurred, it obtains using each region as the candidate of item, and therefrom determines that support is greater than the k- item collection of minimum support threshold value as frequent item set, wherein k >=2;Corresponding candidate association rule is generated according to each frequent item set, and therefrom determines that confidence level is greater than the credible correlation rule of minimal confidence threshold;According to the population situation of preceding paragraph corresponding region in part or all of credible correlation rule, the population situation of corresponding consequent corresponding region is predicted, and the strategy of deploying to ensure effective monitoring and control of illegal activities of management resource is adjusted according to prediction result.This method can according to each region personnel's behavioral characteristics in advance to management Resource allocation and smoothing, deploy to ensure effective monitoring and control of illegal activities, avoid management resource waste.
Description
Technical field
The present invention relates to field of computer technology, in particular to city personnel's compartmentalization cloth keyholed back plate based on multi-source big data
Manage method and system.
Background technique
City refers to the area of denser population, would generally include residential quarter, industrial area and shopping centre etc. in city by making
With the region of property is different, function is different building and facility composition.Someone will flow out into people are for work of going to work in region
The purpose of work, amusement and recreation, can be moved to another region from a region, form the stream of people in the process of moving.
Can also be equipped with management resource in city, managing resource includes to provide public service etc. and city being assisted to carry out normally
The resource for operating and needing to configure, such as sanitarian environmentally friendly personnel are engaged in, safeguard safety inspector, the traffic dispersion of public safety
Member, security patrol person etc., for the current limliting fence for limit and being oriented to the stream of people and be arranged, the police kiosk set up to enhance public security
Deng.
Since management resource is provided to ensure city personnel normal life, management resource is needed in densely populated place
Region between deployed to ensure effective monitoring and control of illegal activities on path or increase the dynamics of deploying to ensure effective monitoring and control of illegal activities, avoid the occurrence of configuration management inadequate resource and cause to have in time
Effect ground is managed personnel;And can reduce the dynamics of deploying to ensure effective monitoring and control of illegal activities on path between the sparse region of personnel, it prevents excessively
The wasting problem of configuration management resource.It follows that the population between the population dynamic and region in each region becomes
Changing relevance can reflect which region needs to strengthen management the deploying to ensure effective monitoring and control of illegal activities of resource to a certain extent.
However, be not at present dynamically changeable to the mode of deploying to ensure effective monitoring and control of illegal activities of management resource, it can not be according to the dynamic of regional population
Variation is deployed to ensure effective monitoring and control of illegal activities to carry out the resource of instantaneity, therefore, need a kind of population change association that can be established between urban area and with
The method that management resource is rationally deployed to ensure effective monitoring and control of illegal activities and deployed based on this.
Summary of the invention
(1) goal of the invention
To overcome above-mentioned at least one defect of the existing technology, waste management resource is avoided, enables and manages resource
Enough to keep efficient use, the present invention provides following technical schemes.
(2) technical solution
As the first aspect of the present invention, the invention discloses a kind of city personnel's compartmentalization cloth based on multi-source big data
Control management method, comprising:
The population development data of multiple regions are obtained from multi-source data;
According to the population development data, the same target for coming across more than one region and each described same is determined
The region that target occurred is obtained using each region as the candidate of item, and therefrom determines that support is greater than minimum support
The k- item collection of threshold value is as frequent item set, wherein k >=2;
Corresponding candidate association rule is generated according to each frequent item set, and therefrom determines that confidence level is set greater than minimum
The credible correlation rule of confidence threshold;
According to the population situation of preceding paragraph corresponding region in the part or all of credible correlation rule, prediction is corresponding consequent right
The population situation in region is answered, and adjusts the strategy of deploying to ensure effective monitoring and control of illegal activities of management resource according to prediction result.
It is to determine to come across in the setting period when determining the same target in a kind of possible embodiment
The same target in more than one region.
In a kind of possible embodiment, the duration of the setting period is determined according at least one of following: the mankind are living
Dynamic and work and rest time interval, the character of use in the region.
It is to determine to come across setting space model when determining the same target in a kind of possible embodiment
Enclose interior same target, wherein the setting spatial dimension includes more than one region.
In a kind of possible embodiment, the size of the minimum support threshold value occurs according to each same target
The quantity in the region crossed is set.
In a kind of possible embodiment, determine that support is greater than minimum support threshold value from the candidate
K- item collection include: as frequent item set
Then the support for calculating each candidate's n- item collection determines that support is more than or equal to the minimum support threshold value of setting
Candidate n- item collection, as frequent n- item collection, wherein n >=1;
Each frequent n- item collection is spliced, candidate (n+1)-item collection is obtained;
It brings obtained candidate into above-mentioned calculating step to be iterated, until only determining a frequent item set or working as
Until secondary all candidates are not frequent item sets;
Frequent k- item collection is filtered out from all obtained frequent item sets, wherein k >=2.
It is described to generate corresponding candidate association rule specifically according to each frequent item set in a kind of possible embodiment
Are as follows:
The data item for extracting each frequent k- item collection determined carries out sequentially group to the data item of same frequent item set
It closes, the data item group for obtaining using each data item or each being combined by multiple data item is preceding paragraph, remainder data Xiang Weihou
The candidate association rule of item.
In a kind of possible embodiment, after determining the frequent k- item collection, count each described k- frequent
The total degree for concentrating each data item to occur screens out the frequent k- item collection for the data item containing total degree less than k of exchanging, and will be remaining
Frequent k- item collection as the frequent item set determined.
In a kind of possible embodiment, after obtaining candidate j- item collection, each data in each candidate's j- item collection are counted
The total degree occurred, screens out the candidate j- item collection for the data item containing total degree less than j of exchanging, and by remaining candidate j-
Collection is as the candidate determined, wherein j >=3.
In a kind of possible embodiment, the strategy of deploying to ensure effective monitoring and control of illegal activities according to prediction result adjustment management resource includes:
The management resource that path is deployed to ensure effective monitoring and control of illegal activities between consequent corresponding region and the corresponding region of preceding paragraph in credible correlation rule with
The same trend increase and decrease of the population situation of the preceding paragraph corresponding region.
In a kind of possible embodiment, it is described according to prediction result adjustment management resource deploy to ensure effective monitoring and control of illegal activities strategy when, also
The strategy of deploying to ensure effective monitoring and control of illegal activities of confidence level adjustment management resource according to corresponding credible correlation rule.
As a second aspect of the invention, city personnel's compartmentalization based on multi-source big data that the invention also discloses a kind of
It deploys to ensure effective monitoring and control of illegal activities management system, comprising:
Demographic data obtains module, for obtaining the population development data of multiple regions from multi-source data;
Same target determination module is determined for obtaining the population development data that module obtains according to the demographic data
The region that the same target and each same target for coming across more than one region out occurred;
Frequent item set determining module, same target and region for being determined according to the same target determination module,
It obtains using each region as the candidate of item, and therefrom determines that support is greater than the k- item collection of minimum support threshold value as frequency
Numerous item collection, wherein k >=2;
Correlation rule determining module, each frequent item set for determining according to the frequent item set determining module generate
Corresponding candidate association rule, and therefrom determine that confidence level is greater than the credible correlation rule of minimal confidence threshold;
Population situation prediction module, for the people according to preceding paragraph corresponding region in the part or all of credible correlation rule
Mouth state predicts the population situation of corresponding consequent corresponding region,
It deploys to ensure effective monitoring and control of illegal activities Developing Tactics module, for the prediction result adjustment management resource according to the population situation prediction module
It deploys to ensure effective monitoring and control of illegal activities strategy.
In a kind of possible embodiment, the same target determination module is true when determining the same target
Make the same target that more than one region is come across in the setting period.
In a kind of possible embodiment, the duration of the setting period is determined according at least one of following: the mankind are living
Dynamic and work and rest time interval, the character of use in the region.
In a kind of possible embodiment, the same target determination module is true when determining the same target
Make the same target come across in setting spatial dimension, wherein the setting spatial dimension includes more than one region.
In a kind of possible embodiment, the size of the minimum support threshold value determines mould according to the same target
The quantity in the region that each same target that block determines occurred is set.
In a kind of possible embodiment, the frequent item set determining module includes:
Then frequent item set determination unit determines that support is greater than for calculating the support of each candidate's n- item collection
In the candidate n- item collection of the minimum support threshold value of setting, as frequent n- item collection, wherein n >=1;
Candidate determination unit, each frequent n- item collection for being determined to the frequent item set determination unit into
Row splicing obtains candidate (n+1)-item collection;
Cycling element, the candidate for obtaining the candidate determination unit are brought the frequent item set into and are determined
Unit, until only determining a frequent item set or until time all candidates are not frequent item sets;
Frequent item set screening unit is used for after the cycling element stops circulation, from all obtained frequent item sets
Frequent k- item collection is filtered out, wherein k >=2.
In a kind of possible embodiment, the correlation rule determining module includes:
Data item extraction unit, for extracting each frequent k- item collection that the frequent item set determining module is determined
Data item;
Data item assembled unit, the data item of the same frequent item set for being extracted to the data item assembled unit into
Row sequentially combines, and the data item group for obtaining using each data item or each being combined by multiple data item is preceding paragraph, its remainder
It is consequent candidate association rule according to item.
In a kind of possible embodiment, the frequent item set determining module further include:
Frequent item set screens out unit, for after the frequent item set determination unit determines the frequent k- item collection,
The total degree that each data item occurs in each frequent k- item collection is counted, the data item exchanged containing total degree less than k is screened out
Frequent k- item collection, and the frequent item set that remaining frequent k- item collection is determined as the frequent item set determination unit.
In a kind of possible embodiment, the frequent item set determining module further include:
Candidate screens out module, for after obtaining candidate j- item collection, counting in the frequent item set determination unit
The total degree that each data item occurs in each candidate's j- item collection screens out candidate j- of the data item exchanged containing total degree less than j
Collect, and the candidate that remaining candidate's j- item collection is determined as the candidate determination unit, wherein j >=3.
In a kind of possible embodiment, the Developing Tactics module of deploying to ensure effective monitoring and control of illegal activities includes:
First Developing Tactics unit, for make in credible correlation rule consequent corresponding region and the corresponding region of preceding paragraph it
Between the management resource deployed to ensure effective monitoring and control of illegal activities of path with the population situation of the preceding paragraph corresponding region same trend increase and decrease.
In a kind of possible embodiment, the Developing Tactics module of deploying to ensure effective monitoring and control of illegal activities further include:
Second Developing Tactics unit, for managing resource according to prediction result adjustment in the first Developing Tactics unit
Deploy to ensure effective monitoring and control of illegal activities strategy when, the also strategy of deploying to ensure effective monitoring and control of illegal activities according to the confidence level adjustment management resource of corresponding credible correlation rule.
(3) beneficial effect
City personnel's compartmentalization provided by the invention based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities management method and system, is had as follows
The utility model has the advantages that
1, personnel's behavioral characteristics according to each region are in advance to managing Resource allocation and smoothing, deploying to ensure effective monitoring and control of illegal activities, so that resource and personnel are dynamic
State feature matches, and can carry out in time to city management resource and reasonably deploy and deploy to ensure effective monitoring and control of illegal activities, avoid the wave of management resource
Take, so that management resource is able to maintain efficient use, achievees the purpose that each takes what he needs, makes the best use of everything.
2, it is divided by the way that situation are occurred in personnel according to the period, preferably population development data can be utilized population
State microcosmicization, and embody personnel's movement law different in each period, the difference for respectively obtaining each period are deployed to ensure effective monitoring and control of illegal activities plan
Slightly, management resource is more fully utilized.
3, by the character of use according to key construction in mankind's activity rule and region, the setting period is arranged
Duration can be more rapidly to people to increase the frequency for implementing prediction when foreseeable population situation pace of change is very fast
Mouth state change is reacted, and then is more rapidly deployed to ensure effective monitoring and control of illegal activities and adjustmenting management resource;It simultaneously can also be in foreseeable population
The frequency for implementing prediction is reduced when state change speed is slower, avoids waste system operations process resource.
4, it is taken by carrying out drawing for setting spatial dimension in the region got, realizes making by oneself for management of implementing to deploy to ensure effective monitoring and control of illegal activities
Adopted regional function.
5, it by the way that the size of minimum support threshold value to be carried out to the floating of anti-trend according to region quantity, can be avoided in area
When domain quantity is more, lead to no frequent item set since the region of same target appearance is more dispersed, and avoid in region
When negligible amounts, since the region that same target occurs more is concentrated and cause to be all frequent item set.
6, by statistical data item number, undesirable frequent item set, candidate is screened out in advance, avoided
The superset as caused by these item collections is calculated, system operations process resource is saved.
Detailed description of the invention
It is exemplary below with reference to the embodiment of attached drawing description, it is intended to for the explanation and illustration present invention, and cannot manage
Solution is the limitation to protection scope of the present invention.
Fig. 1 is that city personnel's compartmentalization provided by the invention based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities management method first embodiment
Flow diagram.
Fig. 2 is each geo-demographics' tables of data.
Fig. 3 is the same object statistics tables of data in each region.
Fig. 4 is each geo-demographics' tables of data of segment mark when having.
Fig. 5 is that city personnel's compartmentalization provided by the invention based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities management system first embodiment
Structural block diagram.
Specific embodiment
To keep the purposes, technical schemes and advantages of the invention implemented clearer, below in conjunction in the embodiment of the present invention
Attached drawing, technical solution in the embodiment of the present invention is further described in more detail.
It should be understood that in the accompanying drawings, from beginning to end same or similar label indicate same or similar element or
Element with the same or similar functions.Described embodiments are some of the embodiments of the present invention, rather than whole implementation
Example, in the absence of conflict, the features in the embodiments and the embodiments of the present application can be combined with each other.Based in the present invention
Embodiment, every other embodiment obtained by those of ordinary skill in the art without making creative efforts,
It shall fall within the protection scope of the present invention.
Herein, " first ", " second " etc. are only used for mutual differentiation, rather than indicate their significance level and sequence
Deng.
The division of module, unit or assembly herein is only a kind of division of logic function, in actual implementation may be used
To there is other division modes, such as multiple modules and/or unit can be combined or are integrated in another system.As separation
The module of part description, unit, component are also possible to indiscrete may be physically separated.It is shown as a unit
Component can be physical unit, may not be physical unit, it can is located at a specific place, may be distributed over grid
In unit.Therefore some or all of units can be selected to realize the scheme of embodiment according to actual needs.
The city personnel's compartmentalization based on multi-source big data provided below with reference to Fig. 1-Fig. 4 the present invention is described in detail is deployed to ensure effective monitoring and control of illegal activities
The first embodiment of management method.The present embodiment is mainly used in urban resource distribution and deploys to ensure effective monitoring and control of illegal activities, can be according to personnel in city
It is distributed the regional regularity presented, according to the behavioral characteristics of personnel in each region, facing area implements deploying to ensure effective monitoring and control of illegal activities for management resource,
Prevent that resource allocation is unreasonable to result in waste of resources and the problem of resource scarcity, resource is made the best use of everything.
As shown in Figure 1, city personnel compartmentalization provided in this embodiment is deployed to ensure effective monitoring and control of illegal activities, management method includes the following steps:
Step 100, the population development data of multiple regions are obtained from multi-source data.
Region refers to that personnel flow into city, the particular space range of outflow and residence, such as a resident society
Area or the residential communities of multiple adjoinings may be constructed a region, a subway station or a railway station or an airport can
To constitute a region, the commercial circle being made of the market of multiple adjoinings or the workspace being made of the office building of multiple adjoinings or
The university city that person is made of more universities also may be constructed a region.Similar can also have park, hotel, gymnasium, public affairs
Public Library etc..It is understood that the multiple regions obtained can be adjacent to each other, other regions are separated between being also possible to
's.
Multi-source data refers to the data got from a variety of different data sources, obtains the channel of data not from each data source
Together, data acquiring mode is also different.For example, the networked door access control system from residential communities/campus entrance obtains resident
Community area/campus region swipes the card/and brush face enters and leaves the information of personnel, from subway station/railway station/airport entrance connection
Gateway machine obtain subway station region/railway station region/aircraft field areas swipe the card/barcode scanning enter and leave personnel information, from store/shopping
Networking monitoring camera or networking MAC acquisition bayonet at the entrance of center obtain store region/shopping center region and enter and leave people
The information of member.Wherein it is possible to by carrying out recognition of face in the picture that shoots to monitoring camera with obtain personal information and
Population situation, and since MAC acquisition bayonet can establish communication connection with everyone portable mobile phone and obtain mobile phone
MAC Address, and the usual people of mobile phone has and one and carries, therefore can pass through the hand that bayonet acquisition is acquired to MAC
Machine MAC Address is recorded and is identified to obtain personal information and population situation.The packet of above-mentioned discrepancy personnel include into
The personnel identity information of personnel.
It is recorded by the discrepancy personal information to each region, and record is summarized, each region can be obtained
Population development data, using as the basic data set rationally deployed to ensure effective monitoring and control of illegal activities to city management resource later.Wherein, population development
Data include that personnel enter and leave quantity (i.e. flow of the people), and respectively enter and leave the identity of personnel.Specifically, being brushed in discrepancy personnel
When card/brush face/barcode scanning/recognition of face/MAC Address identification, identifying system can not only keep a record to floating population number, can also lead to
Cross the identity information that card number/face/barcode scanning application/handset binding etc. identifies discrepancy personnel, with know joint owner's flow with
And whom the target of flowing is respectively, and the region that each target occurred is associated with the target, is obtained as shown in Figure 2 each
Geo-demographics' tables of data.Wherein, black square represents target and is coming across in corresponding region, for the A of region, the stream of people
Amount is 12, and personnel occur includes target 1-10, target 12 and target 14, other regions can similarly obtain.
It should be noted that above-mentioned target refers to all personnel for coming across any region in above-mentioned multiple regions.
In addition, due to acquisition the data scale of construction can more huge (mass data), each data source in multi-source data
The data stored may be big data, and the population development data actually got also can be big data, it is therefore desirable to
It uses the subsequent data mining mode that will be mentioned and carries out data analysis, association identification etc..It is thus understood that and listed by table
The quantity of target is only to illustrate out, does not represent real goal quantity.
Step 200, according to population development data, the same target for coming across more than one region and each same is determined
The region that target occurred is obtained using each region as the candidate of item, and therefrom determines that support is greater than minimum support
The item collection of threshold value is as frequent item set.
After obtaining population development data, the same mesh for coming across more than one region is determined from population development data
Mark.Same target refers to and all targets for coming across more than one region in above-mentioned multiple regions.With the table number in Fig. 2
For, wherein target 11-15 only comes across a region, does not carry out trans-regional shifting in tetra- regions A, B, C, D
It is dynamic, therefore for the management resource in this four regions is deployed to ensure effective monitoring and control of illegal activities, target 11-15 is not deploying to ensure effective monitoring and control of illegal activities what strategy needed to change currently
Aspect generates contribution, is not same target, therefore these factors (target 11-15) are screened out.Determine all same targets
(target 1-10) and then determine therefrom that out the region that each same target occurred, i.e. region A-D.
Item collection is the set of item, and candidate is using each region as the set of item.It determines all same targets and obtains
Behind the region that each same target occurred, it will be able to these Data Integrations be formed candidate, obtain each region as shown in Figure 3
Same object statistics tables of data.Wherein, candidate has { A }, { B }, and { C } and { D } respectively represents the item of region A, B, C and D
Collection.{ A }, { B }, { C } and { D } four item collections are 1- item collection, wherein number 1 represents only one data item in item collection.
The frequency that frequent item set refers to that item collection occurs reaches certain level for all item collections, and judges an item
Whether the frequency of occurrences of collection reaches that become the standard of frequent item set be support and minimum support threshold value.Support is one
The frequency that a item collection occurs, each item collection has the support of oneself, and for the application, the support of item collection is exactly region
The interior frequency for same target occur.Minimum support threshold value is become for judging whether the support of an item collection reaches
The numerical value mark post of frequent item set.By calculating support and using the minimum support threshold value of setting as mark post, from candidate item
Concentration filters out frequent item set.
In one embodiment, determine that support is greater than the k- item collection work of minimum support threshold value from candidate
For frequent item set the following steps are included:
Step 210, then the support for calculating each candidate's n- item collection determines that support is more than or equal to the most ramuscule of setting
The candidate n- item collection of degree of holding threshold value, as frequent n- item collection, wherein n >=1.
By taking the tables of data in Fig. 3 as an example, the support of candidate { A } are as follows: come across the number of the same target in the A of region
Amount/same target total quantity=10/10=1, can similarly obtain, the support of candidate { B }, { C } and { D } is respectively as follows:
0.9,0.7,0.4.The value of minimum support threshold value setting at this time is 0.5, and candidate { D } fails to reach minimum support threshold value
The standard set up, therefore candidate { D } is not frequent item set.And candidate { A }, the support of { B } and { C } are above
Minimum support threshold value, therefore it is confirmed as frequent item set, obtain frequent 1- item collection.
Step 220, each frequent n- item collection is spliced, obtains candidate (n+1)-item collection.
After determining frequent 1- item collection, each frequent 1- item collection is spliced, candidate 2- item collection is obtained.1- item collection is spelled
When being connected in 2- item collection, orderly spliced to each of frequent 1- item collection, obtain n (n-1)/2 candidate's 2- item collection,
Middle n is the quantity of frequent 1- item collection.In the present embodiment, there are three obtained candidate 2- item collections, is respectively { A, B }, { B, C } and
{A,C}。
Step 230, by obtained candidate bring into above-mentioned calculating step be iterated until only determine one frequently
Item collection or until time all candidates are not frequent item sets.
It obtains candidate 2- item collection and then judges whether each candidate 2- item collection is frequent item set.Judgment mode is similarly meter
Then the support for calculating each candidate's 2- item collection judges whether to be higher than minimum support threshold value.Wherein, the minimum support threshold of use
Value is identical as threshold value when judging 1- item collection, is 0.5.Specifically, the support of candidate's 2- item collection { A, B } are as follows: occur
In region A and quantity/same target total quantity=9/10=0.9 of the same target in the B of region is also occurred at, similarly may be used
, the support of item collection { B, C } and { A, C } are respectively 0.6 and 0.7.Since the support of three candidate's 2- item collections is above most
Small support threshold, therefore three candidate's 2- item collections are frequent 2- item collection.
After determining frequent 2- item collection, each frequent 2- item collection is spliced, candidate 3- item collection is obtained.Candidate 3- item collection
It is the item collection of highest item number, only { A, B, C } item collection.
Then judge whether candidate 3- item collection is frequent item set again.Judgment mode is similarly the support for calculating candidate's 3- item collection
Then degree judges whether to be higher than minimum support threshold value.Wherein, the minimum support threshold value of use and threshold when judging 1- item collection
It is identical to be worth numerical value, is 0.5.Specifically, the support of candidate's 3- item collection { A, B, C } are as follows: come across region A, region B and
Quantity/same target total quantity=the 6/10=0.6 for coming across the same target of region C, since the support of { A, B, C } is high
In minimum support threshold value, therefore candidate's 3- item collection { A, B, C } is frequent 3- item collection.Meet at this time and only determines a frequent episode
The condition of collection, therefore the process for finding frequent item set terminates.
Step 240, frequent k- item collection is filtered out from all obtained frequent item sets, wherein k >=2.
{ A } of only one data item, { B }, { C } are screened out, that is to say, that screened out frequent 1- item collection, remaining frequency
Numerous item collection be exactly finally obtain and enter subsequent step frequent item set, comprising: { A, B }, { B, C }, { A, C } and A, B,
C } this four item collections.
Although 1- item collection also has frequent item set, 1- item collection only includes an item, between region after being not used to
Relevance judged, thus while 1- item collection also has a frequent item set, but not counting within the frequent item set referred to later, this
The frequent item set that embodiment hereinafter refers to all refers to the n- item collection of n >=2.
Step 300, corresponding candidate association rule is generated according to each frequent item set, and therefrom determines that confidence level is greater than most
The credible correlation rule of small confidence threshold value.
Correlation rule acts between two item collections, can derive another by an item collection using correlation rule
Collection, obtains the degree of strength of relevance between two item collections.Such as appear in the same target of region A and also appear in region B,
Correlation rule is exactly { A } -> { B }, wherein { A } is known as preceding paragraph, also known as former piece, guide, { B } is referred to as consequent, also known as consequent, subsequent.
In another example the same target for appearing in region B also appears in region A, correlation rule is exactly { B } -> { A }, before { B } is at this time
, { A } is consequent.
Why not 1- item collection is namely covered the reason in frequent item set by this, because correlation rule needs at least two
A data item could be established.
Confidence level obtains the credible/degree of reliability an of conclusion according to some condition, and each frequent item set has oneself
Support confidence level, for the application, the confidence level of item collection is exactly that same target can also be inferred when appearing in certain region
The same target also appears in the confidence level in another region out.Minimal confidence threshold is for judging to be associated between two item collections
Rule whether numerical value mark post believable enough.By calculating confidence level and using the minimal confidence threshold being arranged as mark post,
Credible correlation rule is filtered out from the candidate association rule of each frequent item set.
In one embodiment, the data item for extracting each frequent k- item collection determined, to the number of same frequent item set
Sequentially combined according to item, obtain using each data item or the data item group that is each combined by multiple data item as preceding paragraph,
Remainder data item is consequent candidate association rule.
There are four the frequent item sets obtained by step 200: { A, B }, { B, C }, { A, C } and { A, B, C }, this four frequently
The candidate association rule that item collection generates has region to region: { A } -> { B }, { A } -> { C }, { B } -> { A }, { B } -> { C }, { C }-
> { A } and { C } -> { B }, there are also regions to region group: { A } -> { BC }, { B } -> { AC } and { C } -> { AB }.Wherein, area
Domain group is the entirety of multiple regions composition, is equivalent to data item group.What is obtained at this time is using each data item as preceding paragraph, its remainder
It is the consequent candidate association rule sequentially combined according to item.
Confidence level judgement is carried out to above-mentioned nine candidate association rules, wherein the confidence of candidate association rule { A } -> { B }
Degree are as follows: come across region A and the quantity of the same target that also occurs in the B of region/come across region A same target number
Amount=9/10=0.9, the confidence level of candidate association rule { B } -> { A } are as follows: come across region A and also occur in the B of region
Quantity=9/9=1 of the same target of the quantity of same target/come across region B, can similarly obtain candidate association regular { A } ->
The confidence level of { C }, { B } -> { C }, { C } -> { A } and { C } -> { B } are respectively as follows: 0.6,0.67,1,0.86.
And the confidence level of candidate association rule { A } -> { BC } are as follows: come across region A, region B and also occur at region C
Same target quantity/come across region A same target quantity=6/10=0.6, can similarly obtain candidate association rule
The confidence level of { B } -> { AC } and { C } -> { AB } are respectively 0.67 and 0.86.
The value of minimal confidence threshold setting at this time is 0.7, candidate association rule { A } -> { C }, { B } -> { C }, { A } ->
The confidence level of { BC } and { B } -> { AC } is lower than minimal confidence threshold, illustrates the regions/areas domain group that these correlation rules refer to
Between relevance it is weaker, be not put into credible correlation rule temporarily, thus adjustment city management resource when deploying to ensure effective monitoring and control of illegal activities it is temporary
When do not take into account that the population change factor that occurs between region that these correlation rules represent.And candidate association regular { A } ->
The confidence level of { B }, { B } -> { A }, { C } -> { A }, { C } -> { B } and { C } -> { AB } are higher than minimal confidence threshold, illustrate this
The relevance of population situation is stronger between the regions/areas domain group that a little correlation rules refer to, and belongs to credible correlation rule, therefore adjusting
When deploying to ensure effective monitoring and control of illegal activities of whole city management resource, the population occurred between the region that can be represented by the above-mentioned credible correlation rule determined
Changing factor carries out resource allocation.
In addition, the candidate association rule generated by frequent item set can also include region group to region: { AB } -> { C },
{ BC } -> { A } and { AC } -> { B }.What is obtained at this time is using the data item group that multiple data item form as preceding paragraph, remainder data
Item is the consequent candidate association rule sequentially combined.Confidence of these three region groups to the candidate association rule in region
Degree is respectively as follows: 0.67,1,0.86, and the confidence level of candidate association rule { BC } -> { A } and { AC } -> { B } is higher than min confidence threshold
Value, also belongs to credible correlation rule, therefore can also illustrate population between regions/areas domain group that these credible correlation rules refer to
The relevance of state is stronger.
Step 400, according to the population situation to preceding paragraph corresponding region in part or all of credible correlation rule, prediction is corresponding
The population situation of consequent corresponding region, and the strategy of deploying to ensure effective monitoring and control of illegal activities for managing resource is adjusted according to prediction result.
The credible correlation rule determined usually have it is multiple, and the preceding paragraph that includes in each credible correlation rule and after
Item is also not exactly the same.Continue by taking the citing in step 300 as an example, the credible correlation rule determined in step 300 are as follows: { A }-
> { B }, { B } -> { A }, { C } -> { A }, { C } -> { B }, { C } -> { AB }, { BC } -> { A } and { AC } -> { B } this seven.Wherein include
Preceding paragraph have { A }, { B }, { C }, five kinds of { BC } and { AC }, include it is consequent have { A }, { B } and { AB } these three.
Adjustment management resource strategy of deploying to ensure effective monitoring and control of illegal activities before, first selection adjustment management resource deploy to ensure effective monitoring and control of illegal activities strategy region.Above-mentioned five
In kind of preceding paragraph, it can all deploy to ensure effective monitoring and control of illegal activities the foundation of strategy as adjustment management resource, only can also select a portion as tune
Homogeneous tube reason resource deploy to ensure effective monitoring and control of illegal activities strategy foundation.
After the selected preceding paragraph for needing participation adjustment to deploy to ensure effective monitoring and control of illegal activities, each population situation for participating in preceding paragraph corresponding region is obtained, then evidence
This population situation to predict corresponding consequent corresponding region of each preceding paragraph in credible correlation rule.Population situation includes population
The data such as amount and population development, population development refer to the size of population, composition and the variable condition in Regional Distribution.
In one embodiment, between consequent corresponding region and the corresponding region of preceding paragraph path in credible correlation rule
The management resource deployed to ensure effective monitoring and control of illegal activities with the population situation of the preceding paragraph corresponding region same trend increase and decrease.
By taking preceding paragraph { A } as an example, after credible correlation rule { A } -> { B } has been determined, obtain region A the size of population and
Population variety state, when the size of population increase when, due to preceding paragraph { A } in credible correlation rule it is corresponding it is consequent have it is consequent
{ B }, it can thus be derived that the destination number of region B also can accordingly carry out a degree of growth, because having one in the A of region
The target of fixed number amount is moved to region B as same target.
After the population variety state for predicting region B, such as forecasted population quantity is promoted, then according to prediction knot
Fruit pre-adjust management resource strategy of deploying to ensure effective monitoring and control of illegal activities, by take raising distribute to region B management resource strategy of deploying to ensure effective monitoring and control of illegal activities, in region
It deploys to ensure effective monitoring and control of illegal activities between A and the connection route of region B or some management resources of more deploying to ensure effective monitoring and control of illegal activities, feelings can occur between associated region in this way
When condition, has with happening and deployed and deploy to ensure effective monitoring and control of illegal activities enough management resources in advance come the occurrence of reply.
It should be noted that can not deploy to ensure effective monitoring and control of illegal activities only according to the size of population for monitoring each region and rationally and effectively and manage resource, example
Such as after the size of population for detecting region C rises, can not learn the population increased is which region movement comes from surrounding, because
This simultaneously can not targetedly deploy to ensure effective monitoring and control of illegal activities in respective paths and manage resource.And by monitoring with the related region B's of region C
After the size of population rises, it is known that in the C of region in increased population, partially or almost all from region B (specifically part is gone back
All depending on region C property whether also relevant with other regions), thus can targetedly will management resource deploy to ensure effective monitoring and control of illegal activities to
On path between region B and region C.
Managing resource includes human resources, facility resource, Service Source etc..Wherein, human resources include police strength, security protection people
Member, Security Personnel, traffic dispersion person, security patrol person, sanitationman, garbage reclamation personnel, infra-structure repair personnel et al.
Power resource.Facility resource includes being arranged or cancelling current limliting fence, open up or close stream of people channel, set up interim police kiosk, configuring and patrol
Patrol the schedulable facilities such as vehicle, scheduling mobile public toilet.Service Source includes the driving direction for changing tide lane, adjustment bus
With subway dispatch a car the frequency, adjust electric power dispensing, adjust heating power supply, the adjustment service such as freightways.
When population situation in estimation range increases, human resources can be correspondinglyd increase between associated region
Quantity of deploying to ensure effective monitoring and control of illegal activities, open or add facility resource and corresponding increment and adjusting are carried out to Service Source, to meet more population institutes
The resource distribution needed.Personnel's behavioral characteristics according to each region are in advance to managing Resource allocation and smoothing, deploying to ensure effective monitoring and control of illegal activities, so that resource and personnel
Behavioral characteristics match, and can carry out in time to city management resource and reasonably deploy and deploy to ensure effective monitoring and control of illegal activities, and avoid management resource
Waste achievees the purpose that each takes what he needs, makes the best use of everything so that management resource is able to maintain efficient use.
It in one embodiment, is to determine to come across not in the setting period when determining same target in step 200
The only same target in a region.
After obtaining population development data, the same mesh for coming across more than one region is determined from population development data
Mark, and determining all same target and then determining therefrom that out the region that each same target occurred.It is same determining
During target, the time that can occur by the setting period to target is divided, and will be occurred in the identical period same
One target is divided into together, and the same target occurred in different periods is then divided into the respective period and determines whether to belong to again
In same target.
For the example described in the step 200, be illustrated in figure 4 with when segment mark each geo-demographics' data
Table, in the region A-D statistical data of table shown in Fig. 4, when background represents that m1 assigns to h2 when data are h1 with horizontal line shading
The data of m2 (hereinafter referred to as the first period) at times, and blank background represents when m2 assigns to h3 when data are h2 m3 at times
The data of (hereinafter referred to as the second period), that is to say, that target 1,3,7 comes across region A, and target 4,7,13 comes across region B,
Target 9,10 comes across region C and target 5 comes across region D and betided for the first period, and other targets come across region
Betided for the second period.
And the data of different periods are individually calculated, that is to say, that when determining same target, only according to target at certain
The record for occurring Mr. Yu region in one period is determined.
By taking Fig. 4 as an example, in the first period, the flow of the people of region A is 3, and personnel occur includes target 1, target 3 and target 7,
Region B-D and so on determines what same target and each same target occurred then in the record data of the first period
Region, then successively determine frequent item set and credible correlation rule, finally predict corresponding consequent corresponding area in credible correlation rule
The population situation in domain, if the size of population of preceding paragraph corresponding region rises, the size of population that consequent corresponding region can be predicted also can
Rise, needs to manage resource to deploying to ensure effective monitoring and control of illegal activities on the path between preceding paragraph corresponding region and consequent corresponding region according to prediction result at this time
Or increases to deploy to ensure effective monitoring and control of illegal activities and manage the dynamics of resource.
In second period, the flow of the people of region A is 9, and personnel occur includes target 2, target 4-6, target 8-10, target 12
With target 14, region B-D and so on, then in the record data of the second period, determine same target (target 1-6,8,
10) determine and the region (region A-D) that occurred of each same target, then successively frequent item set ({ A }, { B }, { C }, A,
B }, { B, C }) and credible correlation rule ({ A- > B }, { B- > A }, { C- > B }), it finally predicts corresponding consequent in credible correlation rule
The population situation of corresponding region, if the size of population decline of preceding paragraph corresponding region at this time, can be predicted the people of consequent corresponding region
Mouthful quantity can also decline, and be needed at this time according to prediction result to subtracting on the path between preceding paragraph corresponding region and consequent corresponding region
It deploys to ensure effective monitoring and control of illegal activities less and manages resource.
Other periods are identical as the calculation process of above-mentioned first, second period and management process of deploying to ensure effective monitoring and control of illegal activities.Setting the period can be with
It is set as a hour, for example, 7:00 to 8:00 is a period, 8:00 to 9:00 is next period;Or the setting period
It is set as half an hour, 7:00 to 7:30 is a period, and 7:30 to 8:00 is a period.Front and back is continuous between period, in
Between it is free of discontinuities.Period is also possible to interval, such as is only determined once daily in 7:00 to the same target occurred between 8:00,
And only implement the primary management method of deploying to ensure effective monitoring and control of illegal activities.
It is divided by the way that situation are occurred in personnel according to the period, preferably population development data can be utilized population shape
State microcosmicization, and embody personnel's movement law different in each period, the difference for respectively obtaining each period are deployed to ensure effective monitoring and control of illegal activities strategy,
Management resource is more fully utilized.
It should be noted that if failing to find frequent item set within the setting period, then illustrate the people between current each region
Mouthful state does not have relevance, without credible correlation rule, just deploys to ensure effective monitoring and control of illegal activities strategy without adjusting current management resource yet.
The setting period mentioned hereinabove can be one hour or half an hour etc., but each setting period all immobilizes
's.In order to which the personnel's behavioral characteristics for obtaining operation are more matched with actual conditions, in one embodiment, the period is set
Duration determined according at least one of following: the time interval of mankind's activity and work and rest, the character of use in region.
The time interval of mankind's activity and work and rest refers to the rule of mankind's activity in city, such as working morning peak, evening
Peak and the sleeping time at night etc..And set the period length can according to its relationship mankind's activity between rule and
Variation.
Specifically, assuming that the 7:00 to 9:00 (made using 24 hours carry out expression time below) in the morning is working peak
The 17:00 to 20:00 of phase, afternoon are peak period of coming off duty, and personnel can concentrate within this twice in and out of each region, therefore
In the time interval of peak period on and off duty, the length of period is set as half an hour, that is to say, that per half an hour determines once this
Same target in half an hour finally predicts the population situation of associated region, in foreseeable population shape by operation
Increase the frequency for implementing prediction when state pace of change is very fast, can more rapidly react to population situation variation, in turn
It more rapidly deploys to ensure effective monitoring and control of illegal activities and adjustmenting management resource.
And within night and morning, such as the time interval of evening 23:00 to morning next day 6:00, since mankind's activity is in
Stand-down, population situation will not change substantially, thus the setting period duration in the time interval may be configured as two hours or
Three hours, that is to say, that statistics comes across the same target in more than one region in two hours or three hours, finally pass through operation
Prediction result is obtained, and then adjusts strategy of deploying to ensure effective monitoring and control of illegal activities.It is reduced when foreseeable population situation pace of change is slower and implements prediction
The frequency avoids waste system operations process resource.
Above-mentioned peak period on and off duty and mankind's activity stand-down are removed, can will be set in remaining other time section
The duration of period is set as one hour.
The character of use in region can classify in region according to character of use, such as building and facility can be divided into inhabitation
Building (corresponding residential communities etc.), public building (corresponding market, office building, hospital, gymnasium, cinema, park, subway
Deng), industrial building (corresponding factory etc.) and farm buildings.
For the region in the multiple regions that get based on residential architecture, can come essentially according to human lives' work and rest
The duration of setting period is set, such as above-mentioned peak period on and off duty and mankind's activity are in stand-down.It is more for what is got
Region in a region based on public building, wherein being also corresponded to using market, office building, subway as the building of representative and facility
Above-mentioned peak period on and off duty and mankind's activity are in stand-down, and for as the building of representative and being set using hospital, KTV, bar
Shi Ze may also be someone will come emergency treatment or night to sing when mankind's activity is in stand-down, therefore the area based on hospital, KTV
Domain is not suitable for then being in stand-down using above-mentioned mankind's activity the duration of setting period is arranged, in setting for night and morning
Timing section duration can be set to one hour or one and a half hours.
It should be noted that when the setting period in each region is not exactly the same, according to setting period duration shortest one
Fang Zuowei sets period duration, and the same target in each region is obtained within the setting period.
It is understood that two foundations of above-mentioned setting setting period duration can use simultaneously, common setting setting
The duration of period.
After having got the population development data of multiple regions in step 100, it can determine in step 200 same
Target, but since the region of acquisition may have tens, it is needing to the section for wherein with zone boundary not being division
Domain deploy to ensure effective monitoring and control of illegal activities when managing, if the same target of all areas is all determined, can only use a portion even only
It is the same target data of sub-fraction, therefore in one embodiment, is determining when step 200 determines same target
The same target in setting spatial dimension is come across out, wherein setting spatial dimension includes more than one region.
Deploy to ensure effective monitoring and control of illegal activities to the region to custom field when managing, so that it may set a spatial dimension, and only to this
Target in setting spatial dimension identify and then determines same target.
Setting spatial dimension can be the territorial scope in regular figure, such as need for the subway station week in certain region
Enclose and carry out management of deploying to ensure effective monitoring and control of illegal activities, then using subway station as several kilometers of center of circle radius within range be setting spatial dimension.Set space model
The territorial scope being also possible in irregular figure is enclosed, such as only needs for belonging to three to be substantially in product shape layout area
Three office buildings carry out management of deploying to ensure effective monitoring and control of illegal activities, then three regions comprising these three office buildings are individually marked off to form one and set
Determine spatial dimension.It should be noted that setting spatial dimension at least needs to include two regions, region pass otherwise cannot achieve
Connection.
Spatial dimension further comprises Offshore Units and aerial facility, such as using amusement park as when setting spatial dimension, uses
It can be used as a Sea area in the seagoing taxi of manned visit, and the roller-coaster for playing can be used as one in the air
Region etc..Region in setting spatial dimension can participate in operation and receive management of deploying to ensure effective monitoring and control of illegal activities, if thinking the region of change participative management,
Then it is directly changed setting spatial dimension.
Setting spatial dimension can cooperate the above-mentioned setting period to be used in conjunction with, for example, the population development data packet got
The data in 50 regions are included, merely desire at present to region A-E therein to start between evening 22:00 to carry out when zero on the day of
It deploys to ensure effective monitoring and control of illegal activities management, wherein region A and B is residential communities, and region C and D are market, and region E is hospital, then marks off region A-E
Form setting spatial dimension, and the character of use of the time interval and region A-E according to mankind's activity and work and rest is arranged
The duration of timing section.
In 0:00 between 6:00, due to the facility for having the nights such as hospital to exercise the function, determination in every 1.5 hours is primary
Same target.In 6:00 between 9:00, due to being working peak period, the primary same target of determinations in every 0.5 hour.9:
00 between 11:00, and mobility of people is not high, the primary same target of determination in every 1 hour.In 11:00 between 13:00, due to
It is the dinner hour, might have the outgoing dining of a large amount of personnel, therefore the primary same target of determination in every 0.5 hour.13:00 extremely
Between 17:00, mobility of people is not high, the primary same target of determination in every 1 hour.In 17:00 between 20:00, under being
Class peak period and date for dinner, the primary same target of determination in every 0.5 hour.In 20:00 between 22:00, mobility of people
It is not high, the primary same target of determination in every 1 hour.
Therefore on the day of zero when to same target being determined between evening 22:00 altogether 28 times, wherein producing how many times frequency
Numerous item collection and credible correlation rule, then will how many time adjustment management resource strategy of deploying to ensure effective monitoring and control of illegal activities, otherwise without adjustment.
The quantity in the region as corresponding to the population development data obtained from multi-source data may be more, it is also possible to obtain
The negligible amounts in region corresponding to the population development data got, or the quantity in region corresponding to the population development data of acquisition
The many but later period only carries out the confirmation of same target by setting spatial dimension to wherein sub-fraction region.It follows that
Implement this deploy to ensure effective monitoring and control of illegal activities management method when, based on region quantity have have it is few.And when facing different region quantities, use is identical
Minimum support threshold value to may result in the frequent item set determined when there are many region quantity few, or even can not determine
Any frequent item set out.Therefore in one embodiment, the size of minimum support threshold value occurred according to each same target
The quantity in region set.
When quantity in the region that same target occurred is more, the region that same target occurs may be more dispersed,
Slightly minimum support threshold value can be lowered a bit at this time, such as lower 10%.And in the region that same target occurred
When negligible amounts, the region that same target occurs can more be concentrated, and can slightly be raised minimum support threshold value a bit at this time,
Such as up-regulation 10%.It is can be avoided in this way when region quantity is more, is led since the region of same target appearance is more dispersed
No frequent item set is caused, and is avoided when region quantity is less, since the region that same target occurs more is concentrated and is caused
It is all frequent item set.
Therefore, minimum support threshold value can be set to the increasing for quantity in the region that same target occurred and
It reduces.
In one embodiment, after determining frequent k- item collection in step 230, count each in each frequent k- item collection
The total degree that data item occurs screens out the frequent k- item collection for the data item containing total degree less than k of exchanging, and will be remaining frequent
K- item collection is as the frequent item set determined, wherein k >=2.
For example, being obtained after determining frequent 2- item collection (k=2 at this time) in each candidate's 2- item collection in step 230
{ A, B }, { A, C }, { A, E }, { B, D } and { B, E } this five frequent item sets.Then each data item in this five frequent item sets is gone out
Existing total degree is counted, and obtains data item A-E and occurs 3 times, 3 times, 1 time, 1 time, 2 times respectively, wherein data item C and D
The number of appearance be less than item collection item number 2, therefore screen out exchange the frequent item set containing data item C and D, obtain { A, B }, A,
E } and { B, E } three frequent 2- item collections.Then again these three frequent 2- item collections are carried out the subsequent step such as splicing.
Reason for this is that if the number that frequently certain data item occurs in k- item collection shows this frequently lower than item time k
After being spliced into candidate (k+1)-item collection, which is likely to because being screened out lower than minimum support threshold value item collection
Fall, therefore exclude the item collection in advance, to avoid again calculating superset caused by the item collection in subsequent process, saves
System operations process resource.
For the frequent item set of the frequent 3- item collection and more Gao Xiangci determined from candidate 3- item collection, item is utilized
Number is screened, to screen out undesirable frequent item set.
In one embodiment, after obtaining out candidate's j- item collection in step 230, count each in each candidate's j- item collection
The total degree that data item occurs screens out the candidate j- item collection for the data item containing total degree less than j of exchanging, and by remaining candidate
J- item collection is as the candidate determined, wherein j >=3.
To candidate screen out with above-mentioned screening out similarly to frequent item set, be all to be sieved by data item number
It removes, purpose is also provided to exclude in advance can be avoided by the item collection that can be equally excluded to by item in the next steps
Superset caused by collecting is calculated, and system operations process resource is saved.
When filtering out credible correlation rule from each correlation rule using confidence level, each credible correlation rule for filtering out
Confidence level may to have height to have low, and the difference of height of confidence level is away from may be larger, and the credible correlation rule of different confidence levels
In the management resource that is subject to of corresponding preceding paragraph corresponding region and consequent corresponding region attention degree of deploying to ensure effective monitoring and control of illegal activities be identical.Similarly,
The confidence level of two credible correlation rules is close, but the support of corresponding frequent item set may differ by it is larger, that is to say, that it is practical
The number difference of flowing is more, but at this time in two credible correlation rules corresponding preceding paragraph corresponding region and consequent corresponding region by
To management resource attention degree of deploying to ensure effective monitoring and control of illegal activities be also identical.Deploying to ensure effective monitoring and control of illegal activities and deploying for ZOOM analysis resource, and in order to optimize
Deploy to ensure effective monitoring and control of illegal activities scheme, in one embodiment, according to prediction result adjustment management resource deploy to ensure effective monitoring and control of illegal activities strategy when, also according to accordingly may be used
Believe the strategy of deploying to ensure effective monitoring and control of illegal activities of the confidence level adjustment management resource of correlation rule.
In abovementioned steps 300, the value of minimal confidence threshold setting is 0.7.If calculating credible association in step 300
The confidence level of regular X is 0.72, and the confidence level of credible correlation rule Y is 0.98, then illustrates in two credible correlation rules
In the identical situation of the population dynamic of preceding paragraph corresponding region, the preceding paragraph corresponding region of credible correlation rule Y and consequent correspondence
Population moving amount between region may will be more than between the preceding paragraph corresponding region and consequent corresponding region of credible correlation rule X
Population moving amount.At this time can the preceding paragraph corresponding region of the two credible correlation rules the size of population simultaneously rise when
(assuming that the preceding paragraph of the two credible correlation rules and consequent be all different), preceding paragraph corresponding region to credible correlation rule Y and
Deploy to ensure effective monitoring and control of illegal activities on path between consequent corresponding region manage resource dynamics be greater than credible correlation rule X preceding paragraph corresponding region and after
It deploys to ensure effective monitoring and control of illegal activities on path between item corresponding region and manages the dynamics of resource.
The city personnel's compartmentalization based on multi-source big data provided below with reference to Fig. 5 the present invention is described in detail is deployed to ensure effective monitoring and control of illegal activities management
The first embodiment of system.The present embodiment is mainly used in urban resource distribution and deploys to ensure effective monitoring and control of illegal activities, and can be distributed according to personnel in city
The regional regularity of presentation, according to the behavioral characteristics of personnel in each region, facing area implements deploying to ensure effective monitoring and control of illegal activities for management resource, prevents
Occur that resource allocation is unreasonable to result in waste of resources and the problem of resource scarcity, resource is made the best use of everything.
The present embodiment is the system for implementing above-mentioned management method first embodiment of deploying to ensure effective monitoring and control of illegal activities.As shown in figure 5, the present embodiment
The management system of deploying to ensure effective monitoring and control of illegal activities provided mainly includes that demographic data obtains module, same target determination module, frequent item set determine mould
Block, correlation rule determining module, population situation prediction module and Developing Tactics module of deploying to ensure effective monitoring and control of illegal activities.
Demographic data obtains module for obtaining the population development data of multiple regions from multi-source data.
Same target determination module obtains module with demographic data and connect, for obtaining what module obtained according to demographic data
Population development data determine the region that the same target for coming across more than one region and each same target occurred.
Frequent item set determining module is connect with same target determination module, for determining according to same target determination module
Same target and region, obtain using each region as the candidate of item, and therefrom determine support be greater than minimum support
The k- item collection of threshold value is as frequent item set, wherein k >=2.
Correlation rule determining module is connect with frequent item set determining module, for what is determined according to frequent item set determining module
Each frequent item set generates corresponding candidate association rule, and therefrom determine that confidence level is greater than minimal confidence threshold can gateway
Connection rule.
Population situation prediction module is connect with correlation rule determining module, for according to part or all of credible correlation rule
The population situation of middle preceding paragraph corresponding region predicts the population situation of corresponding consequent corresponding region,
Developing Tactics module of deploying to ensure effective monitoring and control of illegal activities is connect with population situation prediction module, for the prediction according to population situation prediction module
As a result the strategy of deploying to ensure effective monitoring and control of illegal activities of adjustment management resource.
In one embodiment, same target determination module when determining same target, is determined in the setting period
Come across the same target in more than one region.
In one embodiment, the duration for setting the period is determined according at least one of following: mankind's activity and work and rest
Time interval, the character of use in region.
In one embodiment, same target determination module is to determine to come across setting when determining same target
Same target in spatial dimension, wherein setting spatial dimension includes more than one region.
In one embodiment, the size of minimum support threshold value determines each same according to same target determination module
The quantity in the region that target occurred is set.
In one embodiment, frequent item set determining module includes: frequent item set determination unit, the determining list of candidate
Member, cycling element and frequent item set screening unit.
Frequent item set determination unit is used to calculate the support of each candidate's n- item collection, then determines that support is more than or equal to
The candidate n- item collection of the minimum support threshold value of setting, as frequent n- item collection, wherein n >=1.
Candidate determination unit is connect with frequent item set determination unit, for what is determined to frequent item set determination unit
Each frequent n- item collection is spliced, and candidate (n+1)-item collection is obtained.
Cycling element is connect with candidate determination unit and frequent item set determination unit, for determining candidate
The candidate that unit obtains brings frequent item set determination unit into, until only determining a frequent item set or when secondary all times
Until set of choices is not frequent item set.
Frequent item set screening unit connect with cycling element, for being obtained after cycling element stops circulation from all
Frequent k- item collection is filtered out in frequent item set, wherein k >=2.
In one embodiment, correlation rule determining module includes: data item extraction unit and data item assembled unit.
Data item extraction unit is used to extract the data item for each frequent k- item collection that frequent item set determining module is determined.
Data item assembled unit is connect with data item extraction unit, the same frequency for extracting to data item assembled unit
The data item of numerous item collection is sequentially combined, and the data item with each data item or each to be combined by multiple data item is obtained
Group is preceding paragraph, remainder data item is consequent candidate association rule.
In one embodiment, frequent item set determining module further include: frequent item set screens out unit.
Frequent item set screens out unit and connect with frequent item set determination unit, for determining frequency in frequent item set determination unit
After numerous k- item collection, the total degree that each data item occurs in each frequent k- item collection is counted, screens out and exchanges containing total degree less than k's
The frequent k- item collection of data item, and the frequent item set that remaining frequent k- item collection is determined as frequent item set determination unit.
In one embodiment, frequent item set determining module further include: candidate screens out module.
Candidate screens out module and frequent item set determination unit and connection, for obtaining in frequent item set determination unit
To after candidate j- item collection, the total degree that each data item occurs in each candidate's j- item collection is counted, it is small containing total degree to screen out switch
In the candidate j- item collection of the data item of j, and the candidate that remaining candidate's j- item collection is determined as candidate determination unit
Item collection, wherein j >=3.
In one embodiment, Developing Tactics module of deploying to ensure effective monitoring and control of illegal activities includes: the first Developing Tactics unit.
First Developing Tactics unit for make in credible correlation rule consequent corresponding region and the corresponding region of preceding paragraph it
Between the management resource deployed to ensure effective monitoring and control of illegal activities of path with the population situation of preceding paragraph corresponding region same trend increase and decrease.
In one embodiment, it deploys to ensure effective monitoring and control of illegal activities Developing Tactics module further include: the second Developing Tactics unit.
Second Developing Tactics unit is connect with the first Developing Tactics unit, in the first Developing Tactics unit foundation prediction
As a result adjustment management resource deploy to ensure effective monitoring and control of illegal activities strategy when, also deploying to ensure effective monitoring and control of illegal activities according to the confidence level adjustment management resource of corresponding credible correlation rule
Strategy.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by those familiar with the art, all answers
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be with the scope of protection of the claims
It is quasi-.
Claims (10)
- The management method 1. a kind of city personnel's compartmentalization based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities characterized by comprisingThe population development data of multiple regions are obtained from multi-source data;According to the population development data, the same target for coming across more than one region and each same target are determined The region occurred is obtained using each region as the candidate of item, and therefrom determines that support is greater than minimum support threshold value K- item collection as frequent item set, wherein k >=2;Corresponding candidate association rule is generated according to each frequent item set, and therefrom determines that confidence level is greater than min confidence The credible correlation rule of threshold value;According to the population situation of preceding paragraph corresponding region in the part or all of credible correlation rule, corresponding consequent corresponding area is predicted The population situation in domain, and the strategy of deploying to ensure effective monitoring and control of illegal activities for managing resource is adjusted according to prediction result.
- 2. management method of deploying to ensure effective monitoring and control of illegal activities as described in claim 1, which is characterized in that be to determine when determining the same target The same target in more than one region is come across in the setting period.
- 3. management method of deploying to ensure effective monitoring and control of illegal activities as claimed in claim 1 or 2, which is characterized in that be determining when determining the same target The same target in setting spatial dimension is come across out, wherein the setting spatial dimension includes more than one region.
- 4. management method of deploying to ensure effective monitoring and control of illegal activities as described in claim 1, which is characterized in that determine that support is big from the candidate Include: as frequent item set in the k- item collection of minimum support threshold valueThen the support for calculating each candidate's n- item collection determines that support is more than or equal to the time of the minimum support threshold value of setting N- item collection is selected, as frequent n- item collection, wherein n >=1;Each frequent n- item collection is spliced, candidate (n+1)-item collection is obtained;It brings obtained candidate into above-mentioned calculating step to be iterated, until only determining a frequent item set or when time institute Candidate be not frequent item set until;Frequent k- item collection is filtered out from all obtained frequent item sets, wherein k >=2.
- 5. management method of deploying to ensure effective monitoring and control of illegal activities as described in claim 1, which is characterized in that described to manage resource according to prediction result adjustment Strategy of deploying to ensure effective monitoring and control of illegal activities includes:The management resource that path is deployed to ensure effective monitoring and control of illegal activities between consequent corresponding region and the corresponding region of preceding paragraph in credible correlation rule is with described The same trend increase and decrease of the population situation of preceding paragraph corresponding region.
- The management system 6. a kind of city personnel's compartmentalization based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities characterized by comprisingDemographic data obtains module, for obtaining the population development data of multiple regions from multi-source data;Same target determination module is determined out for obtaining the population development data that module obtains according to the demographic data Now in the region that the same target in more than one region and each same target occurred;Frequent item set determining module, same target and region for determining according to the same target determination module, obtains Using each region as the candidate of item, and therefrom determine that support is greater than the k- item collection of minimum support threshold value as frequent episode Collect, wherein k >=2;Correlation rule determining module, each frequent item set for determining according to the frequent item set determining module generate corresponding Candidate association rule, and therefrom determine confidence level be greater than minimal confidence threshold credible correlation rule;Population situation prediction module, for the population shape according to preceding paragraph corresponding region in the part or all of credible correlation rule State predicts the population situation of corresponding consequent corresponding region,It deploys to ensure effective monitoring and control of illegal activities Developing Tactics module, for deploying to ensure effective monitoring and control of illegal activities for the prediction result adjustment management resource according to the population situation prediction module Strategy.
- 7. deploying to ensure effective monitoring and control of illegal activities management system as claimed in claim 6, which is characterized in that the same target determination module is described in the determination It is the same target for determining to come across more than one region in the setting period when same target.
- 8. management system of deploying to ensure effective monitoring and control of illegal activities as claimed in claims 6 or 7, which is characterized in that the same target determination module is in determination It is to determine to come across the same target in setting spatial dimension, wherein the setting spatial dimension packet when the same target Include more than one region.
- 9. deploying to ensure effective monitoring and control of illegal activities management system as claimed in claim 6, which is characterized in that the frequent item set determining module includes:Then frequent item set determination unit is determined that support is more than or equal to and is set for calculating the support of each candidate's n- item collection The candidate n- item collection of fixed minimum support threshold value, as frequent n- item collection, wherein n >=1;Candidate determination unit, for being spelled to each frequent n- item collection that the frequent item set determination unit is determined It connects, obtains candidate (n+1)-item collection;Cycling element, the candidate for obtaining the candidate determination unit bring the frequent item set into and determine list Member, until only determining a frequent item set or until time all candidates are not frequent item sets;Frequent item set screening unit, for being screened from all obtained frequent item sets after the cycling element stops circulation Frequent k- item collection out, wherein k >=2.
- 10. deploying to ensure effective monitoring and control of illegal activities management system as claimed in claim 6, which is characterized in that the Developing Tactics module of deploying to ensure effective monitoring and control of illegal activities includes:First Developing Tactics unit, for making in credible correlation rule, between consequent corresponding region and the corresponding region of preceding paragraph road The management resource that diameter is deployed to ensure effective monitoring and control of illegal activities with the population situation of the preceding paragraph corresponding region same trend increase and decrease.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910087940.1A CN109902934B (en) | 2019-01-29 | 2019-01-29 | Regional city personnel control management method and system based on multi-source big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910087940.1A CN109902934B (en) | 2019-01-29 | 2019-01-29 | Regional city personnel control management method and system based on multi-source big data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109902934A true CN109902934A (en) | 2019-06-18 |
CN109902934B CN109902934B (en) | 2020-09-11 |
Family
ID=66944513
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910087940.1A Active CN109902934B (en) | 2019-01-29 | 2019-01-29 | Regional city personnel control management method and system based on multi-source big data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109902934B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110705819A (en) * | 2019-08-20 | 2020-01-17 | 重庆特斯联智慧科技股份有限公司 | Community resource matching method and system based on big data |
CN110909759A (en) * | 2019-10-10 | 2020-03-24 | 重庆特斯联智慧科技股份有限公司 | Urban area hierarchical management system and method based on population big data |
CN111352171A (en) * | 2020-03-30 | 2020-06-30 | 重庆特斯联智慧科技股份有限公司 | Method and system for realizing artificial intelligence regional shielding security inspection |
CN112115185A (en) * | 2020-09-29 | 2020-12-22 | 广州瀚信通信科技股份有限公司 | Population big data analysis method, system, equipment and medium for government affair law enforcement |
CN112235727A (en) * | 2020-09-02 | 2021-01-15 | 武汉烽火众智数字技术有限责任公司 | Personnel flow monitoring and analyzing method and system based on MAC data |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110225194A1 (en) * | 2010-03-09 | 2011-09-15 | Electronics And Telecommunications Research Institute | Apparatus and method for analyzing information about floating population |
US9536146B2 (en) * | 2011-12-21 | 2017-01-03 | Microsoft Technology Licensing, Llc | Determine spatiotemporal causal interactions in data |
CN106504162A (en) * | 2016-10-14 | 2017-03-15 | 北京锐安科技有限公司 | Same pedestrian's association analysis method and device based on station MAC scan datas |
CN106844585A (en) * | 2017-01-10 | 2017-06-13 | 广东精规划信息科技股份有限公司 | A kind of time-space relationship analysis system based on multi-source Internet of Things location aware |
CN107609107A (en) * | 2017-09-13 | 2018-01-19 | 大连理工大学 | A kind of trip co-occurrence phenomenon visual analysis method based on multi-source Urban Data |
-
2019
- 2019-01-29 CN CN201910087940.1A patent/CN109902934B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110225194A1 (en) * | 2010-03-09 | 2011-09-15 | Electronics And Telecommunications Research Institute | Apparatus and method for analyzing information about floating population |
US9536146B2 (en) * | 2011-12-21 | 2017-01-03 | Microsoft Technology Licensing, Llc | Determine spatiotemporal causal interactions in data |
CN106504162A (en) * | 2016-10-14 | 2017-03-15 | 北京锐安科技有限公司 | Same pedestrian's association analysis method and device based on station MAC scan datas |
CN106844585A (en) * | 2017-01-10 | 2017-06-13 | 广东精规划信息科技股份有限公司 | A kind of time-space relationship analysis system based on multi-source Internet of Things location aware |
CN107609107A (en) * | 2017-09-13 | 2018-01-19 | 大连理工大学 | A kind of trip co-occurrence phenomenon visual analysis method based on multi-source Urban Data |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110705819A (en) * | 2019-08-20 | 2020-01-17 | 重庆特斯联智慧科技股份有限公司 | Community resource matching method and system based on big data |
CN110909759A (en) * | 2019-10-10 | 2020-03-24 | 重庆特斯联智慧科技股份有限公司 | Urban area hierarchical management system and method based on population big data |
CN111352171A (en) * | 2020-03-30 | 2020-06-30 | 重庆特斯联智慧科技股份有限公司 | Method and system for realizing artificial intelligence regional shielding security inspection |
CN111352171B (en) * | 2020-03-30 | 2023-01-24 | 重庆特斯联智慧科技股份有限公司 | Method and system for realizing artificial intelligence regional shielding security inspection |
CN112235727A (en) * | 2020-09-02 | 2021-01-15 | 武汉烽火众智数字技术有限责任公司 | Personnel flow monitoring and analyzing method and system based on MAC data |
CN112115185A (en) * | 2020-09-29 | 2020-12-22 | 广州瀚信通信科技股份有限公司 | Population big data analysis method, system, equipment and medium for government affair law enforcement |
Also Published As
Publication number | Publication date |
---|---|
CN109902934B (en) | 2020-09-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109902934A (en) | City personnel's compartmentalization based on multi-source big data is deployed to ensure effective monitoring and control of illegal activities management method and system | |
CN110876126B (en) | City control management system based on wisdom street lamp | |
CN103195279B (en) | Based on the intelligent stereo garage of distributed wireless networks | |
CN108198416A (en) | A kind of mobile phone signaling and the fusion method of road network big data and its application and system | |
CN107120583A (en) | The street lamp facility and implementation method of collection illumination, automobile charging and radio communication function | |
CN106576383A (en) | Deployment criteria for unmanned aerial vehicles to improve cellular phone communications | |
CN103280121A (en) | Intelligent parking lot system based on internet of things technology and work method thereof | |
CN104735648B (en) | A kind of pseudo-base station interference analysis and monitoring method and device | |
CN109902930B (en) | Auxiliary system for planning matched facilities based on real population indexes | |
CN106060465A (en) | Dynamic target tracking system and dynamic target tracking method based on electronic map and video surveillance | |
CN107170283A (en) | A kind of parking stall decision-making technique and system | |
CN109949606A (en) | A kind of big data analysis method of intelligent parking | |
US20180114437A1 (en) | System of correlated video and lighting for parking management and control | |
CN110545558A (en) | High-speed service area crowd density estimation system based on Wi-Fi data | |
CN107622415A (en) | Remote control and advertisement delivery system and method based on classification garbage can | |
Reis et al. | Smarter cities with parked cars as roadside units | |
CN110516866A (en) | A kind of real-time estimation method for handing over subway crowding for city rail | |
CN110378552A (en) | Community security personal scheduling system and method based on gridding | |
CN103441911A (en) | Intelligent parking management method based on SNMP, intelligent parking system and gateway | |
CN104240522A (en) | Self-adaptive crossroad control technology based on vehicle area network and fuzzy neural network | |
CN113284024A (en) | Intelligent emergency management system for intensive personnel place | |
CN110033613B (en) | Smart city management method and system based on regional traffic synchronism | |
CN115866547A (en) | Fixed area tourist counting method, system and storage medium based on signaling data | |
Haldorai et al. | Intelligent spectrum handovers in cognitive radio networks | |
CN107991953A (en) | One kind building cental system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |