CN113254718B - Query relaxation method for semantic association search on graph data - Google Patents
Query relaxation method for semantic association search on graph data Download PDFInfo
- Publication number
- CN113254718B CN113254718B CN202010089733.2A CN202010089733A CN113254718B CN 113254718 B CN113254718 B CN 113254718B CN 202010089733 A CN202010089733 A CN 202010089733A CN 113254718 B CN113254718 B CN 113254718B
- Authority
- CN
- China
- Prior art keywords
- query
- entity
- entities
- priority
- tuple
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
A query relaxation method for semantic association searching on graph data, comprising the steps of: giving entity association diagrams and diameter constraints, inputting a group of query entities, respectively calculating the priorities of the query entities, adding the tuples of the < entity, the initial entity and the priority > into a priority queue, taking out a queue head tuple as long as the priority queue is not empty, verifying the query entities, calculating the maximum successful sub-query set which meets the distance condition, and updating the optimal solution; if the priority of the current entity cannot obtain a better solution, terminating and completing the query. The invention solves the problem that the entity association search result is empty under the condition of the given diameter on the graph data, loosens the query entity, and ensures that the result sub-query can find the association.
Description
Technical Field
The invention belongs to the technical field of computers, relates to a graph searching technology, and discloses a query relaxation method for semantic association searching on graph data.
Background
The graph data has good expressive power, such as RDF data, and can intuitively display complex relationships between entities, namely points on the graph, and is applied to more and more fields. In particular, the graph data is well suited for answering associative queries. That is, given several query entities on a graph, the query result is a connected sub-graph that can contain the several query entities, the entities of the sub-graph representation and their relationships are referred to as semantic associations.
There are many search techniques on the graph, comparing basic depth-first searches, breadth-first searches, and variations of both searches. In dealing with the problems associated with graph searching, both methods are often disjunct.
In general, the graph data is a finite directed graph, also called entity association graph, denoted as g=<E,A,R,l>. Where E is a set of entities, denoted as vertices in the figure; a is an arc set, and the direction of each arc a epsilon A points to a head node h (a) from a tail node t (a), t (a) epsilon E and h (a) epsilon E; r is a relationship set; A.fwdarw.R represents that l marks the arc a (a.epsilon.A) and the relation l (a). Epsilon.R. For a given set of non-empty querying entitiesIf a semantic association x=<E x ,A x >Is composed of vertex E x And arc A x Is satisfied, it satisfies the following conditions:
(1) Its vertices contain all query entities, i.e
(2) It is connected;
(3) It is minimal, i.e., neither subset of it can meet both conditions (1) (2) above.
From the above conditions, it is easy to derive that the semantic association corresponds to a tree and that the leaf nodes are all query entities. The diameter of a semantic association is the maximum distance (number of hops) between any two entities.
The prior art generally requires a compact structure when searching for semantic associations, such as limiting the number of entities in the semantic association result or the diameter of the subgraph. A compact association often reveals a tighter and more meaningful relationship between entities, which also meets the needs of the user, but this causes a problem that after limiting the size of the semantic association, it may not be possible to find sub-graphs connecting all query entities, so that the results given by these methods may be empty, which creates a bad experience for the user. To avoid the generation of empty results, query relaxation is required for semantically related searches.
Query relaxation is a technique that can increase the number of information retrieval results. For dealing with situations where the result is empty if the constraints are too high or too tight when making a query. Query relaxation is precisely for this case, where constraints are relaxed appropriately so that the result is no longer empty. For example, when a user uses a search engine, too many keywords may result in too few or even no search results, and the keywords are appropriately pruned, so that the search results are often increased, and the search effect is improved. Limiting the size of semantic associations on graph data, when the search results are empty, the situation is similar, and deleting several query entities appropriately ensures that the remaining query entities can find associations.
Disclosure of Invention
The invention aims to solve the problems that: aiming at the condition that the semantic association search result on the graph data is empty, the association query needs to be relaxed, so that the relaxed sub-query can search for the association.
The technical scheme of the invention is as follows: a query relaxation method for semantic association search on graph data regards semantic association as a tree, leaf nodes in the tree are all entities, wherein the entity used for searching is a query entity, the diameter of the semantic association is the maximum distance between any two entities, and for a given entity association graph G and diameter constraint D and a non-empty query entity set Q, the maximum successful sub-query set Q of Q is calculated max The method is used for searching and realizing query relaxation;
calculate Q max When the method is used, firstly, the priority pr of all query entities qe is calculated respectively, and a tuple is established<qe,qe,pr>Adding the query entity qe of the queue head tuple into a priority queue, taking out the queue head tuple as long as the priority queue is not empty, namely, the tuple with the highest priority in the current priority queue, verifying the query entity qe of the queue head tuple, taking the query entity qe of the queue head tuple as a verification point, and calculating all query entities meeting distance conditions relative to the verification point to obtain a sub-query set Q e As the optimal solution, Q calculated as the current tuple e Q greater than the preceding group of tuples e Updating the optimal solution until the priority of the current entity cannot obtain a better solution, and terminating to obtain the maximum Q e I.e. the maximum successful sub-query set Q max The method comprises the steps of carrying out a first treatment on the surface of the The distance condition is as follows:
(1) The distance from the query entity in any sub-query set to the verification point is not more than
(2) If the diameter constraint is odd, then the distance to the verification point is for allIs the query entity of (1)The verification point has a neighbor point, and the distance from the neighbor point to the query entities is +.>
Preferably, the method further comprises the steps of:
(1) An empty priority queue pq is initially established, the priority queue is a maximum heap, and the stored elements are tuples<Entity, originating entity, priority>Is marked as<e,sqe,pr>，Q max An empty set is initially used for recording the current optimal solution, a closed set is set for recording the verified entities, firstly, for each query entity qe, the priority pr is calculated, and a tuple is obtained<qe,qe,pr>Add pq and access list visited at entity qe Qe, for e, the priority pr is calculated as follows:
pr(e)＝|{sqe}∪{qe∈(Q{sqe}):dist(e,sqe)+dist(e,qe′)≤D}|
wherein sqe represents the initial query entity of e, all query entities in the tuples are queried from the initial query entity, namely, the initial query entity is taken as a verification point, dist represents the distance between two entities, qe' is other query entities except e, and I·| represents the number of set elements;
(2) If pq is not null, executing (3), otherwise turning to (9);
(3) The first tuple is fetched from pq, i.e. the highest priority tuple in the current priority queue, if pr of this tuple does not exceed |Q max I, or less than 2, go to (9), otherwise continue execution (4);
(4) If e in the fetched tuple is not in the closed set, namely, e is not verified, continuing to execute (5), otherwise, turning to (7);
(5) Verifying e, wherein the verification result is Q e ；
(6) Adding the verified e to the locked set if Q e Ratio Q max Containing more elements, i.e. |Q e |>|Q max I, update Q max Has a value of Q e Continuing to execute (7);
(7) If the distance e to sqe is less thanAnd pr is greater than |Q max I, then for all neighbors e 'of e, if e' is not accessing the list visited qe And e ' is accessed along the shortest path from sqe, the priority pr ' of e ' is calculated and the tuple is calculated<e′,sqe,pr′>Adding e' to the virtual in pq qe In (a) and (b);
(8) Turning to the step (2);
(9) Return Q max 。
Further, the step (5) specifically comprises:
(5.1) calculating a distance to e not exceedingQuery entity set Q of (1) 1 And the distance to e is exactly +.>Query entity set Q of (1) 2 ；
(5.2) if Q 1 Ratio Q max Contains more elements, continues execution (5.3), otherwise go to (5.6);
(5.3) if D is even, or Q 2 Containing no more than 1 querying entity, then Q 1 Give Q e Turning to (5.5), otherwise executing (5.4);
(5.4) find Q for all neighbors of e 2 The query entity distance in (1) isThe most numerous neighbors e' of (a), which corresponds to the (2) th of the distance conditions, the corresponding set of querying entities is R e′ And R is taken as e′ ∪(Q 1 \Q 2 ) Give Q e ；
(5.5) if Q e If the size of (1) exceeds 1, Q is returned e Otherwise, turning to (5.6);
(5.6) returning to the empty set.
To ensure that the resulting sub-queries are as close as possible to the original query targets, a minimum number of query entities are deleted when relaxation is required. Specifically, the technical scheme of the invention is how to calculate the maximum successful sub-query set Q of Q for a given entity association graph G, diameter constraint D and an association query Q max Maximum successful sub-query set Q max I.e., a collection of query entities that are capable of implementing a query, corresponding to the deletion of a minimum number of query entities, thereby implementing query relaxation.
The beneficial effects of the invention are as follows: aiming at the situation that semantic association search on graph data is empty, no solution is provided for how to relax a query entity set of association search in the existing query relaxation scheme, the invention firstly provides an effective query relaxation method for semantic association search, in order to ensure consistency with the original query as much as possible, the invention provides a scheme for deleting the least query entity to obtain the largest successful sub-query set, and the invention does not need to find out exact semantic association to ensure the correctness of a relaxation result, but can indirectly ensure through two distance conditions, thereby enhancing the expandability.
Drawings
Fig. 1 is an overall process flow diagram of the present invention.
Fig. 2 is a flow chart of entity verification of the present invention.
FIG. 3 is an exemplary diagram of an associative query in accordance with the methods of the present invention.
Detailed Description
For the diameter D of the limited associated query result, the invention provides a query relaxation method based on the best preferential search of the rapid distance calculation, and finds a maximum successful sub-query set of the associated query set, so that the empty result is avoided during the associated search. The method does not need to search out the actual association, but verifies the existence of the association through a verification point. Usually, whether a result of a query needs to be found out to determine an exact semantic association is verified, but in the invention, a related query can be indirectly judged by a verification point to be successful, and only a certain distance condition is required to be met between the verification point and the query entity, specifically:
(1) The distance from the query entity in any sub-query to the verification point is not more than
(2) If the diameter is odd, then the distance to the verification point is for allThe key query entities of the verification point has a neighbor, and the distance between the neighbor point and the query entities is +.>
Condition (2) is to prevent the situation where the restriction diameter D is actually associated with d+1 when it is odd, that is, if there is a common edge in the path of the key querying entity to the verification point, this situation can be avoided. In order to reduce the search range, the present invention employs a best-priority search. Specifically, a priority is calculated for each entity, each entity may prove the success of a sub-query as a verification point, and the entity with the higher priority will be verified first. When the result is not likely to be better, the algorithm will terminate.
As shown in fig. 1 and 2, the present invention is embodied as follows.
(1) An empty priority queue pq is initially established, which is a maximum heap. Setting the maximum successful sub-query set Q max And the initial empty set is used for updating and recording the current optimal result. An entity whose secured set record has been authenticated is set. For each query entity qe, calculate the priority pr and group the tuples<qe,qe,pr>Adding pq, accessing list of the initial entity qe Qe was added to the mixture. The meaning of a tuple is<Entity, originating entity, priority>For convenience of distinction, it is noted as<e,sqe,pr>For e, its priority pr is calculated as follows:
pr(e)＝|{sqe}∪{qe∈(Q\{sqe}):dist(e,sqe)+dist(e,qe’)≤D}|。
wherein sqe represents the initial query entity of e, all query entities in the tuples are queried from the initial query entity, namely, the initial query entity is taken as a verification point, dist represents the distance between the two entities, and qe' is other query entities except e; for < qe, qe, pr >, the query entity qe is e, also sqe;
(2) Continuing to execute the step (3) if pq is not null, otherwise turning to the step (9);
(3) The first tuple is fetched from pq, i.e. the tuple with the highest priority in the current priority queue. If pr of the tuple does not exceed |Q max I, or less than or equal to 1, turning to (9), otherwise, continuing to execute (4);
(4) If the query entity qe in the fetched tuple is not in the closed set, i.e. qe is not verified, continuing to execute (5), otherwise turning to (7);
(5) Verifying e, wherein the verification result is Q e ：
(5.1) calculating the distance to e not exceedingQuery entity set Q of (1) 1 And a distance to e of exactly +.>Query entity set Q of (1) 2 ；
(5.2) if Q 1 Elements greater than Q max Continuing to execute the step (5.3), otherwise turning to the step (5.6);
(5.3) if D is even, or Q 2 Containing no more than 1 querying entity, then Q 1 Give Q e Turning to (5.5), otherwise executing (5.4);
(5.4) find Q for all neighbors of e 2 The query entity distance in (1) isThe most numerous neighbors e' of (a), which corresponds to the (2) th of the distance conditions, the corresponding set of querying entities is R e′ And R is taken as e′ ∪(Q 1 \Q 2 ) Give Q e ；
(5.5) if Q e If the size of (1) exceeds 1, Q is returned e Otherwise, turning to (5.6);
(5.6) returning to the empty set.
(6) The verified e is added to the closed set. If the current calculated Q e Ratio Q max Containing more elements, i.e. |Q e |>|Q max I, update Q max Is Q e Continuing to execute (7);
(7) If the distance e to sqe is less thanAnd pr is greater than |Q max I, then for all neighbors e 'of e, if e' is not accessing the list visited sqe And e ' is accessed along the shortest path from sqe, the priority pr ' of e ' is calculated and the tuple is calculated<e′,sqe,pr′>Adding e' to the virtual in pq sqe In (a) and (b); e to sqe may also be equal toThe neighbors are not extended any more;
(8) Rotating (2);
(9) Return Q max 。
The present invention is further described in conjunction with the accompanying drawings and detailed embodiments to enable one skilled in the art to practice the invention in light of the description.
Although entity association graphs are directed, they are often considered undirected graphs in semantic association searches, and thus the direction of arcs in the graphs is not considered by the present invention.
The complete flow chart of the invention is shown in fig. 1, and the flow of verifying an entity is shown in fig. 2. The method mainly comprises the following steps: and (3) giving an entity association diagram and a diameter, firstly calculating the priority of a query entity, then adding the tuples of the < entity, the initial entity and the priority > into a priority queue, taking out the first tuple of the queue as long as the priority queue is not empty, verifying the entity, calculating the maximum successful sub-query of which the distance condition is met, and updating the optimal solution. The method may terminate if the priority of the current entity has failed to yield a better solution.
Fig. 3 is an example of an entity association diagram. Rounded rectangles represent entities and arrowed lines represent the relationship between two entities. Wherein the query entities are marked with black and white words { Alice, bob, dan, gary }, other Carol, erin, frank, paper, paper02, paper03 and ISWC2019 are other non-query entities on the graph data, paper represents an article, ISWC2019 is a meeting, other is a person, arrow line author represents an author, such as Paper01 is Alice, is accepted at represents that Paper has been received by ISWC2019 meeting, member of PC represents Dan is a member of PC, and supervisor represents a supervisor, such as Gary's supervisor is Frank. Let the constraint diameter be 4, for the query entity set { Alice, bob, dan, gary }, from the entity association graph, it is apparent that there is no semantic association between the four query entities that satisfies the constraint, since Gary is more than 4 from the other three query entities. If the semantic association search is directly performed, an empty set is generated, if the query is relaxed, gary is removed, and the rest three query entities can find a semantic association with the diameter of 4, so that an empty result is avoided, and therefore, the query is required to be relaxed. First, four tuples are obtained to be inserted into the priority queue pq:
t1＝<Alice,Alice,3>，
t2＝<Bob,Bob,4>，
t3＝<Dan,Dan,4>，
t4＝<Gary,Gary,3>。
taking Alice's priority calculation as an example, alice's priority is 3 because:
dist(Alice,Alice)+dist(Alice,Bob)＝4≤4，
dist(Alice,Alice)+dist(Alice,Dan)＝3≤4，
pr(Alice)＝|{Alice,Bob,Dan}|＝3。
dist (Alice) +dist (Alice, gary) = 6>4, so Gary is not included in the priority calculation of Alice.
Priority calculations for other entities are the same. In pq, t2 (or t 3) has the highest priority of 4, dequeues first, but Bob is validated and found to be unable to validate the success of any sub-query. Then extend Bob, access his neighbors, a new tuple is inserted into pq:
t5＝<Paper02,Bob,4>。
then t5 (or t 3) dequeues. Paper02 was verified and found to prove success of { Bob, dan, gary }, which was assigned to Q max . The neighbors of Paper02 are accessed, and three new tuples are added to pq:
t6＝<ISWC2019,Bob,3>，
t7＝<Erin,Bob,1>，
t8＝<Frank,Bob,2>。
then t3 dequeues. Dan was verified and found to be unable to prove the success of any sub-queries. Accessing Dan's neighbors, a new tuple is added to pq:
t9＝<ISWC2019,Dan,4>。
note that ISWC2019 is enqueued twice, but is accessed from different querying entities. t9 dequeues, validating ISWC2019, finding that it cannot prove the ratio Q max Larger sub-queries succeed. Accessing its neighbors, three new tuples join pq:
t10＝<Paper01,Dan,2>，
t11＝<Paper02,Dan,3>，
t12＝<Paper03,Dan,1>。
the maximum priority in pq is now 3, not exceeding |Q max The value of i. The algorithm terminates and returns the result Alice, bob, dan. { Alice, bob, dan } is the resulting maximum successful sub-query set Q max The query relaxation of the query entity set { Alice, bob, dan, gary } is achieved.
Claims (2)
1. A query relaxation method for semantic association search on graph data is characterized in that semantic association is regarded as a tree, leaf nodes in the tree are entities, wherein the entity used for searching is a query entity, the diameter of the semantic association is the maximum distance between any two entities, and the query is given toThe determined entity association graph G and the diameter constraint D, and a non-empty query entity set Q, calculate the maximum successful sub-query set Q of Q max The method is used for searching and realizing query relaxation;
calculate Q max When the method is used, firstly, the priority pr of all query entities qe is calculated respectively, and a tuple is established<qe,qe,pr>Adding the query entity qe of the queue head tuple into a priority queue, taking out the queue head tuple as long as the priority queue is not empty, namely, the tuple with the highest priority in the current priority queue, verifying the query entity qe of the queue head tuple, taking the query entity qe of the queue head tuple as a verification point, and calculating all query entities meeting distance conditions relative to the verification point to obtain a sub-query set Q e As the optimal solution, Q calculated as the current tuple e Q greater than the preceding group of tuples e Updating the optimal solution until the priority of the current entity cannot obtain a better solution, and terminating to obtain the maximum Q e I.e. the maximum successful sub-query set Q max The method comprises the steps of carrying out a first treatment on the surface of the The distance condition is as follows:
(1) The distance from the query entity in any sub-query set to the verification point is not more than
(2) If the diameter constraint is odd, then the distance to the verification point is for allThe verification point has a neighbor point, and the distance between the neighbor point and the query entities is +.>
Calculate Q max The method comprises the following steps:
(1) An empty priority queue pq is initially established, the priority queue is a maximum heap, and the stored elements are tuples<Entity, originating entity, priority>Is marked as<e,sqe,pr>，Q max An empty set is initially used for recording the current optimal solution, and a closed set is setFor recording entities that have been verified, first for each querying entity qe, the priority pr is calculated, resulting in a tuple<qe,qe,pr>Add pq and access list visited at entity qe Qe, for e, the priority pr is calculated as follows:
pr(e)＝|{sqe}∪{qe∈(Q\{sqe}):dist(e,sqe)+dist(e,qe′)≤D}|
wherein sqe represents the initial query entity of e, all query entities in the tuples are queried from the initial query entity, namely, the initial query entity is taken as a verification point, dist represents the distance between two entities, qe' is other query entities except e, and I·| represents the number of set elements;
(2) If pq is not null, executing (3), otherwise turning to (9);
(3) The first tuple is fetched from pq, i.e. the highest priority tuple in the current priority queue, if pr of this tuple does not exceed |Q max I, or less than 2, go to (9), otherwise continue execution (4);
(4) If e in the fetched tuple is not in the closed set, namely, e is not verified, continuing to execute (5), otherwise, turning to (7);
(5) Verifying e, wherein the verification result is Q e ；
(6) Adding the verified e to the locked set if Q e Ratio Q max Containing more elements, i.e. |Q e |>|Q max I, update Q max Has a value of Q e Continuing to execute (7);
(7) If the distance e to sqe is less thanAnd pr is greater than |Q max I, then for all neighbors e' of e, if e ′ Not accessing list visited qe And e ′ Is accessed along the shortest path from sqe, e is calculated ′ And the tuple is assigned to the priority pr' of (1)<e′,sqe,pr′>Added to pq, e ′ Adding visited qe In (a) and (b);
(8) Turning to the step (2);
(9) Return Q max 。
2. The query relaxation method of semantic association search on graph data according to claim 1, wherein the step (5) specifically comprises:
(5.1) calculating a distance to e not exceedingQuery entity set Q of (1) 1 And the distance to e is exactly +.>Query entity set Q of (1) 2 ；
(5.2) if Q 1 Ratio Q max Contains more elements, continues execution (5.3), otherwise go to (5.6);
(5.3) if D is even, or Q 2 Containing no more than 1 querying entity, then Q 1 Give Q e Turning to (5.5), otherwise executing (5.4);
(5.4) find Q for all neighbors of e 2 The query entity distance in (1) isThe most numerous neighbors e' of (a), which corresponds to the (2) th of the distance conditions, the corresponding set of querying entities is R e′ And R is taken as e′ ∪(Q 1 \Q 2 ) Give Q e ；
(5.5) if Q e If the size of (1) exceeds 1, Q is returned e Otherwise, turning to (5.6);
(5.6) returning to the empty set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010089733.2A CN113254718B (en) | 2020-02-13 | 2020-02-13 | Query relaxation method for semantic association search on graph data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010089733.2A CN113254718B (en) | 2020-02-13 | 2020-02-13 | Query relaxation method for semantic association search on graph data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113254718A CN113254718A (en) | 2021-08-13 |
| CN113254718B true CN113254718B (en) | 2023-08-29 |
Family
ID=77219731
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010089733.2A Active CN113254718B (en) | 2020-02-13 | 2020-02-13 | Query relaxation method for semantic association search on graph data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113254718B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6535632B1 (en) * | 1998-12-18 | 2003-03-18 | University Of Washington | Image processing in HSI color space using adaptive noise filtering |
| CN104408070A (en) * | 2014-10-31 | 2015-03-11 | 北京邮电大学 | Similar sub-image inquiring method and system for protecting privacy under cloud computing environment |
| CN107451210A (en) * | 2017-07-13 | 2017-12-08 | 北京航空航天大学 | A kind of figure matching inquiry method based on inquiry relaxation result enhancing |
| CN109992786A (en) * | 2019-04-09 | 2019-07-09 | 杭州电子科技大学 | A kind of semantic sensitive RDF knowledge mapping approximate enquiring method |
| CN110569368A (en) * | 2019-09-12 | 2019-12-13 | 南京大学 | Query relaxation method for questions and answers of RDF knowledge base |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8301633B2 (en) * | 2007-10-01 | 2012-10-30 | Palo Alto Research Center Incorporated | System and method for semantic search |
| WO2017018901A1 (en) * | 2015-07-24 | 2017-02-02 | Oracle International Corporation | Visually exploring and analyzing event streams |
-
2020
- 2020-02-13 CN CN202010089733.2A patent/CN113254718B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6535632B1 (en) * | 1998-12-18 | 2003-03-18 | University Of Washington | Image processing in HSI color space using adaptive noise filtering |
| CN104408070A (en) * | 2014-10-31 | 2015-03-11 | 北京邮电大学 | Similar sub-image inquiring method and system for protecting privacy under cloud computing environment |
| CN107451210A (en) * | 2017-07-13 | 2017-12-08 | 北京航空航天大学 | A kind of figure matching inquiry method based on inquiry relaxation result enhancing |
| CN109992786A (en) * | 2019-04-09 | 2019-07-09 | 杭州电子科技大学 | A kind of semantic sensitive RDF knowledge mapping approximate enquiring method |
| CN110569368A (en) * | 2019-09-12 | 2019-12-13 | 南京大学 | Query relaxation method for questions and answers of RDF knowledge base |
Non-Patent Citations (1)
| Title |
|---|
| 基于本体的XML近似查询方法研究;杨宏娜;《中国优秀硕士学位论文全文数据库信息科技辑》(第5期);第I138-1341页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113254718A (en) | 2021-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ngo et al. | Worst-case optimal join algorithms | |
| Yang et al. | Cost-based variable-length-gram selection for string collections to support approximate queries efficiently | |
| Yu et al. | Keyword search in relational databases: A survey. | |
| US7814042B2 (en) | Selecting candidate queries | |
| Mottin et al. | Exemplar queries: Give me an example of what you need | |
| Xiao et al. | Ed-join: an efficient algorithm for similarity joins with edit distance constraints | |
| Loekito et al. | A binary decision diagram based approach for mining frequent subsequences | |
| Xiao et al. | Top-k set similarity joins | |
| US8396852B2 (en) | Evaluating execution plan changes after a wakeup threshold time | |
| Kolaitis et al. | Efficient querying of inconsistent databases with binary integer programming | |
| Koutris et al. | The data complexity of consistent query answering for self-join-free conjunctive queries under primary key constraints | |
| Semertzidis et al. | Top-$ k $ Durable Graph Pattern Queries on Temporal Graphs | |
| Zhou et al. | Fast SLCA and ELCA computation for XML keyword queries based on set intersection | |
| Hadjieleftheriou et al. | Hashed samples: selectivity estimators for set similarity selection queries | |
| Lu et al. | Efficiently Supporting Edit Distance Based String Similarity Search Using B $^+ $-Trees | |
| JP2005267612A (en) | Improved query optimizer using implied predicates | |
| Hong et al. | Subgraph matching with set similarity in a large graph database | |
| Mouratidis et al. | Joint search by social and spatial proximity | |
| Li et al. | Supporting search-as-you-type using sql in databases | |
| Park et al. | Efficient processing of keyword queries over graph databases for finding effective answers | |
| Choi et al. | Skyline queries on keyword-matched data | |
| Zhang et al. | Modeling and computing probabilistic skyline on incomplete data | |
| Anadiotis et al. | Integrating connection search in graph queries | |
| Das et al. | Duplicate reduction in graph mining: Approaches, analysis, and evaluation | |
| CN113254718B (en) | Query relaxation method for semantic association search on graph data |
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 |