CN105912574A - Multi-user decided spatial data query and verification method - Google Patents

Abstract

The invention discloses a multi-user decided spatial data query and verification method comprising a spatial data processing step S1, a query data giving step S2, a query initialization step S3, a data query step S4, and a data verification step S5. The multi-user decided spatial data query and verification method provided by the invention can be applied to a multi-user decided scenario. Users in a group each get a good result according to the location and hobbies thereof. An effective, complete and correct method is provided for multi-user based large-amount spatial data query and verification in a big data environment.

Description

The Spatial data query verification method that a kind of multi-user determines

Technical field

The present invention relates to the Spatial data query verification method that a kind of multi-user determines.

Background technology

Along with based on venue services universal, various tissues, company even individual's (being referred to as DOs) starts to collect and have Substantial amounts of spatial data；Processing such substantial amounts of spatial data is that data owner brings technology and economically huge is chosen War；In the management and query script of data, DOs tends to data are contracted out to third party rather than set up themselves Technical team and basic framework；Because alleviating DOs great burden in data management processes, the potential offer of data outsourcing More effective and cost-effective service；In data sub-contract management, the data of oneself are delegated to ISP (SP), SP by DO Index, the search request of feedback user is set up for data；Because server is not in the range of the administration authority of DO, server is permissible Distorting the Query Result of return, therefore, user is necessary to ensure that Query Result meets three conditions: effectiveness, correctness, completely Property.

Number of patent application: CN201510101056.0 discloses a kind of space querying integrity based on Merkle tree construction Verification method, the method is on four points of tree nodes that existing adaptive H ilbert curve is generated, it is proposed that support to have inquired about The construction method of the Merkle tree construction of integrity verification, and propose the integrity verification method of range query and KNN inquiry, make Obtain the situation that integrity verification result provided by the present invention does not exist wrong report and fails to report, and then it is right that ISP is difficult to The Query Result of user carries out malice to be changed；The inventive method can carry for user under the service mode of spatial data outsourcing For efficient checking structural generation function, and true scope inquiry inquires about integrity verification function with KNN, thus ensures sky Between inquire about the quality of service.Only provide the space querying integrity verification method of a kind of Merkle tree construction, data are had Effect property and correctness are not correlated with and are related to, and are not carried out determining multi-user the feedback of inquiry yet.

Number of patent application: CN201310132565.0 discloses a kind of data dynamic operation verifiability based on Hash tree Method, is to be connected by communication network by user USER, cloud computation data center CDC and auditing by third party mechanism TPA tri-part Composition；Proposition one side that USER asks as data storage service, it is desirable to one's own data file is stored cloud computing Among the cloud storage space of data center；USER both can be personal user, it is also possible to be enterprise customer；CDC is responsible for response and uses The data storage service request at family, stores oneself huge data center according to certain rule by the data file of user, And the management of data file safeguarded be responsible for；TPA is as reliable auditing by third party mechanism, by the trust of USER to being stored in The data file of CDC data center carries out integrity and conforming examination；The present invention solve under cloud computing environment for User data file integrality and conforming validation problem；Only provide the integrity to data and conforming checking, not There is the feedback realizing that multi-user is determined inquiry.

Existing data query method such as k-Nearest Neighbors inquiry and Skyline inquiry all can not be applied to In the scene that multi-user determines, the user of i.e. one group can not obtain a best result according to respective place with hobby.

Summary of the invention

It is an object of the invention to overcome the deficiencies in the prior art, it is provided that the Spatial data query that a kind of multi-user determines is tested Card method, it is possible to being applied in the scene that multi-user determines, the user of a group obtains one according to respective place and hobby Well result, Query Result possesses effectively, completely, correct advantage.

It is an object of the invention to be achieved through the following technical solutions: the Spatial data query that a kind of multi-user determines is tested Card method, comprises the following steps:

S1. spatial data handling: the site objects of spatial data is grouped according to space attribute, each packet conduct One external square of minimum, forms MR-tree using the external square of each minimum as a leaf node, generates correspondence for each leaf node Index and store corresponding numerical value；MR-tree is a kind of Merkle Hash tree, and it includes multiple intermediate node, intermediate node In one or more leaf nodes of comprising, these leaf nodes are all its child nodes；

S2. inquiry data give: given quaternary array Q={U, and W, P, k}, U represent user's group of inquiry, and P represents user The place set of group；W represents user and defines hobby set, and k is the site objects quantity that user needs；

U={u₀,u₁,...u_i...,u_n-1, u in formula_i, representing that user organizes i+1 user in U, n represents that user organizes in U The number of user,

P={p₀,p₁,...,p_i,...,p_s, p in formula_i, i=1,2,3.....s, represent i-th place in place set P Object, s represents the number of site objects in place set P,

W={w₀,w₁,...,w_i...,w_n-1, w in formula_i, i=1,2,3.....n-1, represent the user-defined love of i-th Good；

w_i={ w_i,o,w_i,1,...,w_i,m, w_i,oThe space attribute weight of hobby, { w is defined for i-th user_i,1,..., w_i,mIt it is the i-th user non-spatial attributes weight that defines hobby.

It is defined as follows concept:

Non-space is arranged: given two place p and p ', if p ' is poor unlike p on all non-spatial attributes, then specifies P ' non-space domination p, symbolization is:

Space is arranged: a given user gathers U ' and two place p and p ', if p ' non-space domination p, and p ' ratio P will be near apart from all users, and we then specify to gather user have p ' to arrange p ' on U '；

Non-space weight is arranged: given the three unities p and the hobby set W of all users, the weight of the non-spatial attributes of p Be expressed as: ◇ (p):

Given two place p and p ', if ◇ (p ') is big unlike ◇ (p), then it is assumed that p ' non-space weight domination p；

Space weight domination: given two place p, the hobby set W of p ' and all users, p is to the power of all users It is heavily:

${\mathrm{ad}}_{sum}\left(p\right)=\underset{i=0}{\overset{n-1}{\Σ}}\left(w_{i0*}\right||p,u_i|\left|\right),$

If ad_sum(p') unlike ad_sumP () is big, it is believed that p ' space weight domination p；

Weight is arranged: a given user collects U, and all of consumer taste collection W and two place p ' and p, if p ' does not props up Join p, p ' non-space weight domination p, and p ' space weight domination p, then it is assumed that collect weight domination p on U user；

The added value of one node: v_sum(p)=ad_sum(p)+◇(p)；

The added value of one scope:

Given Q={U, after W, P, k}, desired result is that final k result is stored in result set R, and meets not by P The domination of other places or weight domination；

S3. initialization is inquired about: define query results R and identifying object collection VO, and by query results R and identifying object Collection VO is initialized as sky；Calculate intermediate node and the leaf node v at place, each place at MR-tree each packet place_sum；

S4. data query: define a heap H and come with v_sumAscending order scan the node of MR-tree: first by MR-tree's Root node is put in heap H and identifying object collection VO, scans each time, is all ejected by the heap top element of heap H, carries out heap top element Check according to checking that result, to identifying object collection VO, query results R and heap H process, until heap H is empty, completes once Basic query, check in R | | the R | | the size with user defined parameters k of counting:

(1) if | | R | | is more than k, by place according to v_sumNumerical values recited sorts, and takes front k and is worth to final result collection R′；

(2) if | | R | | is equal to k, directly using R as final result collection R '；

(3) if | | R | | is less than k, next round inquiry is carried out, until | | R | | is not less than k；

S5. data verification: according to final result collection R ' with during VO, the effectiveness of checking final result collection R ', integrity and Correctness.

The numerical value of each leaf node storage described in step S1 is:

H=hash (p¹|p²|...|pⁱ...|p^t),

P in formulaⁱ, i=1,2,3.....t represent the i-th site objects in the packet of leaf node correspondence, and ' | ' represents over the ground Point object carries out cascade operation, and t represents the number of site objects in this leaf node.

Described MR-tree has multiple intermediate node, and each intermediate node includes one or more leaf node.

If the three unities p domination or weight arrange another place p or area S, must have: v_sum(p) ＜ v_sum(p') or v_sum(p) ＜ v_sum(S), any one v based on place_sumTravel through MR-tree by ascending order, add the ground of candidate result collection R to Point concentrates R ' by certainly putting into final result.

Heap top element is checked by step S4 and carries out respective handling and be divided into following four situation:

(1) if heap top element is arranged by the element in existing R or weight is arranged, heap top element is put in VO；

(2) if heap top element is not arranged by the element in existing R or weight is arranged, and heap top element is MR- The intermediate node of tree, joins its child node in heap H；

(3) if heap top element is not arranged by the element in existing R or weight is arranged, and heap top element is leaf segment Point, all places it comprised join in heap H；

(4) if heap top element is not arranged by the element in existing R or weight is arranged；Heap top element is place pair As, it is just directly placed in R.

Described step S5 includes following sub-step:

S51. check preparation: each take turns inquiry time result set R and corresponding VO carried out being grouped and be ordered as checking make Good prepare, during poll-final, final identifying object collection VO contains three kinds of data: site objects；The number of minimum external square It is worth the intermediate node that square external with minimum is corresponding；Query Information U and W of all users；

S52. validity check: check the numerical value calculating root node according to final identifying object collection VO, then with former number Value compares, it is judged that calculated numerical value is the most identical with former numerical value:

(1) if identical, then R ' possesses effectiveness；

(2) if it is not the same, then R ' do not possess effectiveness；

S53. Correctness checking: each packet G in the place number of inspection final result collection R ' and R '_i, judge R ' Correctness；

S54. integrity checking: check each packet G in R '_iWith VO corresponding in VO_i, judge the integrity of R ', as The really G in R '_iAll complete, then R ' has integrity.

Described step S53 includes following sub-step:

S531. check whether final result collection R ' meets condition one: final result collection R ' place number defines equal to user Parameter k:

(1) if meeting condition one, then step S532 is jumped to；

(2) if being unsatisfactory for condition one, then final result collection R ' does not possess correctness；

S532. each packet G in R ' is checked_i, it is judged that G_iWhether meet condition two: G_iIn place the most mutually arrange, also Not mutually weight domination:

(1) if meeting condition two, then G_iPossess correctness, jump to step S533；

(2) if being unsatisfactory for condition two, then final result collection R ' does not possess correctness:

S533. each site objects in final result collection R ' is checked, it may be judged whether meet condition three: high-grade packet Site objects can be arranged or the site objects of weight domination inferior grade packet:

(1) if meeting condition three, final result collection R ' possesses correctness；

(2) if being unsatisfactory for condition three, final result collection R ' does not possess correctness.

Described step S54 includes following sub-step:

S541. VO is judged_iIn site objects, if be satisfied by condition four: by corresponding G_iIn site objects domination or power Heavily arrange；

(1) if meeting condition four, then corresponding G_iPossesses integrity；

(2) if being unsatisfactory for condition four, step S542 is jumped to；

S542. VO is judged_iIn site objects, if meet condition five: for being unsatisfactory for the site objects of condition four, its Added value is higher than G_iIn the added value of all site objects:

(1) if meeting condition five, then corresponding G_iPossesses integrity；

(2) if being unsatisfactory for condition five, then corresponding G_iDo not possesses integrity.

Spatial data query based on multi-user and checking, make query script then prove that each result is to generate result Effectively, correct and complete, meet needs under big data environment and process the data owner's of high amount spatial data Demand, and propose Spatial data query based on Merkle Hash tree and verification method innovatively, improve inquiry speed Spend and decrease the size returning result set；And under different parameters is arranged highly effective rate and robustness.

The invention has the beneficial effects as follows: for the looking into of spatial data of the big data quantity based on multi-user under big data environment Ask and checking provides a kind of effective, completely, correct method, and have faster than original general data querying method Inquiry velocity, shorter response time and less return result set, decrease the spending of user.

Accompanying drawing explanation

Fig. 1 is the flow chart of the present invention；

Fig. 2 is the scanning process figure during the data query of the present invention；

Fig. 3 is embodiments of the invention one schematic diagrams.

Detailed description of the invention

Technical scheme is described in further detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to The following stated.

As it is shown in figure 1, the Spatial data query verification method that a kind of multi-user determines, comprise the following steps:

S1. spatial data handling: the site objects of spatial data is grouped according to space attribute, each packet conduct One external square of minimum, forms MR-tree using the external square of each minimum as a leaf node, generates correspondence for each leaf node Index and store corresponding numerical value；MR-tree is a kind of Merkle Hash tree, and it includes multiple intermediate node, intermediate node In one or more leaf nodes of comprising, these leaf nodes are all its child nodes；

S2. inquiry data give: given quaternary array Q={U, and W, P, k}, U represent user's group of inquiry, and P represents user The place set of group；W represents user and defines hobby set, and k is the site objects quantity that user needs；

U={u₀,u₁,...u_i...,u_n-1, u in formula_i, representing that user organizes i+1 user in U, n represents that user organizes in U The number of user,

P={p₀,p₁,...,p_i,...,p_s, p in formula_i, i=1,2,3.....s, represent i-th place in place set P Object, s represents the number of site objects in place set P,

W={w₀,w₁,...,w_i...,w_n-1, w in formula_i, i=1,2,3.....n-1, represent the user-defined love of i-th Good；

w_i={ w_i,o,w_i,1,...,w_i,m, w_i,oThe space attribute weight of hobby, { w is defined for i-th user_i,1,..., w_i,mIt it is the i-th user non-spatial attributes weight that defines hobby.

It is defined as follows concept:

Non-space is arranged: given two place p and p ', if p ' is poor unlike p on all non-spatial attributes, then specifies P ' non-space domination p, symbolization is:

Space is arranged: a given user gathers U ' and two place p and p ', if p ' non-space domination p, and p ' ratio P will be near apart from all users, and we then specify to gather user have p ' to arrange p ' on U '；

Non-space weight is arranged: given the three unities p and the hobby set W of all users, the weight of the non-spatial attributes of p Be expressed as: ◇ (p):

Given two place p and p ', if ◇ (p ') is big unlike ◇ (p), then it is assumed that p ' non-space weight domination p；

Space weight domination: given two place p, the hobby set W of p ' and all users, p is to the power of all users It is heavily:

${\mathrm{ad}}_{sum}\left(p\right)=\underset{i=0}{\overset{n-1}{\Σ}}\left(w_{i0*}\right||p,u_i|\left|\right),$

If ad_sum(p') unlike ad_sumP () is big, it is believed that p ' space weight domination p；

Weight is arranged: a given user collects U, and all of consumer taste collection W and two place p ' and p, if p ' does not props up Join p, p ' non-space weight domination p, and p ' space weight domination p, then it is assumed that collect weight domination p on U user；

The added value of one node: v_sum(p)=ad_sum(p)+◇(p)；

The added value of one scope:

Given Q={U, after W, P, k}, desired result is that final k result is stored in result set R, and meets not by P The domination of other places or weight domination；

S3. initialization is inquired about: define query results R and identifying object collection VO, and by query results R and identifying object Collection VO is initialized as sky；Calculate intermediate node and the leaf node v at place, each place at MR-tree each packet place_sum；

S4. data query: define a heap H and come with v_sumAscending order scan the node of MR-tree: as in figure 2 it is shown, first The root node of MR-tree is put in heap H and identifying object collection VO, scan each time, all the heap top element of heap H is ejected, right Heap top element carries out checking according to checking that result, to identifying object collection VO, query results R and heap H process, until heap H is Sky, completes a basic query, check in R | | the R | | the size with user defined parameters k of counting:

(1) if | | R | | is more than k, by place according to v_sumNumerical values recited sorts, and takes front k and is worth to final result collection R′；

(2) if | | R | | is equal to k, directly using R as final result collection R '；

(3) if | | R | | is less than k, next round inquiry is carried out, until | | R | | is not less than k；

S5. data verification: according to final result collection R ' with during VO, the effectiveness of checking final result collection R ', integrity and Correctness.

The numerical value of each leaf node storage described in step S1 is:

H=hash (p¹|p²|...|pⁱ...|p^t),

P in formulaⁱ, i=1,2,3.....t represent the i-th site objects in the packet of leaf node correspondence, and ' | ' represents over the ground Point object carries out cascade operation, and t represents the number of site objects in this leaf node.

Described MR-tree has multiple intermediate node, and each intermediate node includes one or more leaf node.

If the three unities p domination or weight arrange another place p or area S, must have: v_sum(p) ＜ v_sum(p') or v_sum(p) ＜ v_sum(S), any one v based on place_sumTravel through MR-tree by ascending order, add the ground of candidate result collection R to Point will be put in final result collection R ' certainly.

As in figure 2 it is shown, heap top element is checked by step S4 and carries out respective handling and be divided into following four situation:

(1) if heap top element is arranged by the element in existing R or weight is arranged, heap top element is put in VO；

(2) if heap top element is not arranged by the element in existing R or weight is arranged, and heap top element is MR- The intermediate node of tree, joins its child node in heap H；

(3) if heap top element is not arranged by the element in existing R or weight is arranged, and heap top element is leaf segment Point, all places it comprised join in heap H；

(4) if heap top element is not arranged by the element in existing R or weight is arranged；Heap top element is place pair As, it is just directly placed in R.

Described step S5 includes following sub-step:

S51. check preparation: each take turns inquiry time result set R and corresponding VO carried out being grouped and be ordered as checking make Good prepare, during poll-final, final identifying object collection VO contains three kinds of data: site objects；The number of minimum external square It is worth the intermediate node that square external with minimum is corresponding；Query Information U and W of all users；

S52. validity check: check the numerical value calculating root node according to final identifying object collection VO, then with former number Value compares, it is judged that calculated numerical value is the most identical with former numerical value:

(1) if identical, then R ' possesses effectiveness；

(2) if it is not the same, then R ' do not possess effectiveness；

S53. Correctness checking: each packet G in the place number of inspection final result collection R ' and R '_i, judge R ' Correctness；

S54. integrity checking: check each packet G in R '_iWith VO corresponding in VO_i, judge the integrity of R ', as The really G in R '_iAll complete, then R ' has integrity.

Described step S53 includes following sub-step:

S531. check whether final result collection R ' meets condition one: final result collection R ' place number defines equal to user Parameter k:

(1) if meeting condition one, then step S532 is jumped to；

(2) if being unsatisfactory for condition one, then final result collection R ' does not possess correctness；

S532. each packet G in R ' is checked_i, it is judged that G_iWhether meet condition two: G_iIn place the most mutually arrange, also Not mutually weight domination:

(1) if meeting condition two, then G_iPossess correctness, jump to step S533；

(2) if being unsatisfactory for condition two, then final result collection R ' does not possess correctness:

S533. each site objects in final result collection R ' is checked, it may be judged whether meet condition three: high-grade packet Site objects can be arranged or the site objects of weight domination inferior grade packet:

(1) if meeting condition three, final result collection R ' possesses correctness；

(2) if being unsatisfactory for condition three, final result collection R ' does not possess correctness.

Described step S54 includes following sub-step:

S541. VO is judged_iIn site objects, if be satisfied by condition four: by corresponding G_iIn site objects domination or power Heavily arrange；

(1) if meeting condition four, then corresponding G_iPossesses integrity；

(2) if being unsatisfactory for condition four, step S542 is jumped to；

S542. VO is judged_iIn site objects, if meet condition five: for being unsatisfactory for the site objects of condition four, its Added value is higher than G_iIn the added value of all site objects:

(1) if meeting condition five, then corresponding G_iPossesses integrity；

(2) if being unsatisfactory for condition five, then corresponding G_iDo not possesses integrity.

Spatial data query based on multi-user and checking, make query script then prove that each result is to generate result Effectively, correct and complete, meet needs under big data environment and process the data owner's of high amount spatial data Demand, and propose Spatial data query based on Merkle Hash tree and verification method innovatively, improve inquiry speed Spend and decrease the size returning result set；And under different parameters is arranged highly effective rate and robustness.

Embodiment one, as it is shown on figure 3, data owner (Data Owner is called for short DO) by spatial data according to step S1 Carry out being grouped, using each packet as leaf node, carry out the process of Merkle Hash tree foundation, and be that each packet generates one Numerical value, and the data after processing are supplied to third-party ISP (Service Provider, be called for short SP)；When multiplex U, the place of user's group, when ISP initiates data inquiry request, are organized according to the user of request data inquiry in family (User) Set P；User defines hobby set W, site objects quantity k that user needs, and forms quaternary array (quaternary formula) Q={U, W, P, k}, ISP carries out data initialization, data query and data verification, by finally give further according to step S3～S5 Result feeds back to each user, thus multi-user has obtained a good Query Result according to respective place and hobby.

Claims (8)

1. the Spatial data query verification method that a multi-user determines, it is characterised in that: comprise the following steps:

S1. spatial data handling: be grouped the site objects of spatial data according to space attribute, each packet is as one Minimum external square, forms MR-tree using the external square of each minimum as a leaf node, generates corresponding rope for each leaf node Draw and store corresponding numerical value；

S2. inquiry data give: given quaternary array Q={U, and W, P, k}, U represent user's group of inquiry, and P represents user's group Place is gathered；W represents user and defines hobby set, and k is the site objects quantity that user needs；

S3. initialization is inquired about: define query results R and identifying object collection VO, and by query results R and identifying object collection VO It is initialized as sky；Calculate intermediate node and the leaf node v at place, each place at MR-tree each packet place_sum；

S4. data query: define a heap H and come with v_sumAscending order scan the node of MR-tree: first the root of MR-tree is saved Point is put in heap H and identifying object collection VO, scans each time, is all ejected by the heap top element of heap H, checks heap top element According to checking that result, to identifying object collection VO, query results R and heap H process, until heap H is empty, completes the most basic Inquiry, check in R | | the R | | the size with user defined parameters k of counting:

(1) if | | R | | is more than k, by place according to v_sumNumerical values recited sorts, and takes front k and is worth to final result collection R '；

(2) if | | R | | is equal to k, directly using R as final result collection R '；

(3) if | | R | | is less than k, next round inquiry is carried out, until | | R | | is not less than k；

S5. data verification: during according to final result collection R ' and VO, the effectiveness of checking final result collection R ', integrity is with correct Property.

The Spatial data query verification method that a kind of multi-user the most according to claim 1 determines, it is characterised in that: step The numerical value of each leaf node storage described in S1 is:

H=hash (p¹|p²|...|pⁱ...|p^t),

P in formulaⁱ, i=1,2,3.....t represent that the i-th site objects in the packet of leaf node correspondence, ' | ' expression are carried out over the ground Cascade operation, t represents the number of site objects in this leaf node.

The Spatial data query verification method that a kind of multi-user the most according to claim 1 determines, it is characterised in that: described MR-tree there is multiple intermediate node, each intermediate node includes one or more leaf node.

The Spatial data query verification method that a kind of multi-user the most according to claim 1 determines, it is characterised in that: any One v based on place_sumTraveling through MR-tree by ascending order, the place adding candidate result collection R to is final by certainly putting into In result set R '.

The Spatial data query verification method that a kind of multi-user the most according to claim 1 determines, it is characterised in that: step Heap top element is checked by S4 and carries out respective handling and be divided into following four situation:

(1) if heap top element is arranged by the element in existing R or weight is arranged, heap top element is put in VO；

(2) if heap top element is not arranged by the element in existing R or weight is arranged, and heap top element is MR-tree Intermediate node, joins its child node in heap H；

(3) if heap top element is not arranged by the element in existing R or weight is arranged, and heap top element is leaf node, will All places that it comprises join in heap H；

(4) if heap top element is not arranged by the element in existing R or weight is arranged；Heap top element is site objects, will It is just directly placed in R.

The Spatial data query verification method that a kind of multi-user the most according to claim 1 determines, it is characterised in that: described Step S5 include following sub-step:

S51. check preparation: each take turns inquiry time result set R and corresponding VO being carried out is grouped and is ordered as checking and performs standard Standby, during poll-final, final identifying object collection VO contains three kinds of data: site objects；The numerical value of minimum external square and The intermediate node that minimum external square is corresponding；Query Information U and W of all users；

S52. validity check: check the numerical value calculating root node according to final identifying object collection VO, then enter with former numerical value Row compares, it is judged that calculated numerical value is the most identical with former numerical value:

(1) if identical, then R ' possesses effectiveness；

(2) if it is not the same, then R ' do not possess effectiveness；

S53. Correctness checking: each packet G in the place number of inspection final result collection R ' and R '_i, judge that R's ' is correct Property；

S54. integrity checking: check each packet G in R '_iWith VO corresponding in VO_i, judge the integrity of R ', if R ' In G_iAll complete, then R ' has integrity.

The Spatial data query verification method that a kind of multi-user the most according to claim 6 determines, it is characterised in that: described Step S53 include following sub-step:

S531. check whether final result collection R ' meets condition one: final result collection R ' place number is equal to user's defined parameters k:

(1) if meeting condition one, then step S532 is jumped to；

(2) if being unsatisfactory for condition one, then final result collection R ' does not possess correctness；

S532. each packet G in R ' is checked_i, it is judged that G_iWhether meet condition two: G_iIn place the most mutually arrange, the most not phase Weight domination mutually:

(1) if meeting condition two, then G_iPossess correctness, jump to step S533；

(2) if being unsatisfactory for condition two, then final result collection R ' does not possess correctness:

S533. each site objects in final result collection R ' is checked, it may be judged whether meet condition three: the place of high-grade packet Object can be arranged or the site objects of weight domination inferior grade packet:

(1) if meeting condition three, final result collection R ' possesses correctness；

(2) if being unsatisfactory for condition three, final result collection R ' does not possess correctness.

The Spatial data query verification method that a kind of multi-user the most according to claim 6 determines, it is characterised in that: described Step S54 include following sub-step:

S541. VO is judged_iIn site objects, if be satisfied by condition four: by corresponding G_iIn site objects domination or weight prop up Join；

(1) if meeting condition four, then corresponding G_iPossesses integrity；

(2) if being unsatisfactory for condition four, step S542 is jumped to；

S542. VO is judged_iIn site objects, if meet condition five: for being unsatisfactory for the site objects of condition four, its add Value is higher than G_iIn the added value of all site objects:

(1) if meeting condition five, then corresponding G_iPossesses integrity；

(2) if being unsatisfactory for condition five, then corresponding G_iDo not possesses integrity.