CN110059264B - Site retrieval method, equipment and computer storage medium based on knowledge graph - Google Patents
Site retrieval method, equipment and computer storage medium based on knowledge graph Download PDFInfo
- Publication number
- CN110059264B CN110059264B CN201910333290.4A CN201910333290A CN110059264B CN 110059264 B CN110059264 B CN 110059264B CN 201910333290 A CN201910333290 A CN 201910333290A CN 110059264 B CN110059264 B CN 110059264B
- Authority
- CN
- China
- Prior art keywords
- entity
- node
- query
- index
- place
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- 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
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- 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
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a place retrieval method, equipment and a computer storage medium based on a knowledge graph, which comprise an indexing method and a query method, wherein under the influence of two factors of space and semantics on a query result, space and semantic indexes are respectively established and stored in a combined way, a linear quadtree index structure combining space and semantic information is designed, and an efficient query algorithm aiming at the index structure is designed. Aiming at the defect of low query efficiency of the traditional KSP method under the condition of massive data of the RDF undirected graph of the knowledge graph, the invention improves the query efficiency of the online place while guaranteeing the accuracy of the query result. Experiments prove that the method can greatly improve the efficiency of inquiring the knowledge graph on-line places.
Description
Technical Field
The present invention relates to a location retrieval method, apparatus, and computer storage medium, and more particularly, to a location retrieval method, apparatus, and computer storage medium based on a knowledge graph.
Background
Knowledge in the knowledge graph is represented by an RDF data model, the RDF models the data as an entity and establishes connection with other entities or types, namely, a triplet form of a < subject, a predicate and an object >, wherein the subject and the object are both entities, and the predicate represents the relationship of the two entities. Each entity has its own global ID, type and text description, the text description stores the related description and keyword set of the corresponding entity, each node represents an entity in the RDF diagram, and each edge represents the relationship of two nodes.
The query input of the user is a keyword set required by the user, and the query system needs to return a place entity set which is most in line with the semantic requirement of the user. The semantic distance refers to the semantic association degree between the to-be-queried place entity and the actual query requirement of the user, the smaller the semantic distance is, the more accords with the query semantic of the user, and the semantic distance is represented in the RDF graph by the sum of the shortest distance between the to-be-queried entity and the node containing the query keyword of the user. The spatial location is also an important factor affecting the query result of the user, so that the query system needs to combine the spatial location information, and the spatial distance is the euclidean distance between the location of the user and the spatial geographic location of the location entity to be queried.
The KSP inquiry is to find out the location entities close to the user according to the space distance, and to search each location entity in the RDF graph with breadth first, find out the subtree meeting the following conditions (1) the root node is a location entity (2) the keyword input by the user is contained in the node text description in the subtree. If the above conditions are met, the location entity is a location meeting the semantic requirements of the user. And then, comprehensively sequencing the cost functions according to the space distance and the semantic distance, and returning to the Top-k place entity sets.
The traditional KSP queries mainly through RDF directed graphs, cannot completely express the true semantics of users, has large storage and calculation cost under the condition of large data volume, can find that the data graphs need to be used as undirected graphs for query according to the true semantic information, greatly increases the query data volume, so that the traditional KSP method cannot efficiently query the RDF undirected graphs on line, and meanwhile, the cost judgment mode for result ordering is not accurate. Fig. 1 is a simple place query example, where P1 represents "beijing", P2 represents "homeland", V1 represents "national 5A-level scenic spot", where (a) there is an RDF triplet < beijing, including, homeland >, < homeland, belonging to national 5A-level scenic spot >, (b) there is a triplet < national 5A-level scenic spot, including, homeland >, < homeland, and Beijing >, it can be found that (a) and (b) have the same semantics, and when a user is to find the same result of "city including 5A-level scenic spot" or "which city is covered by" 5A-level scenic spot "should return to P1" in the conventional KSP method, the user query requirement cannot be completely satisfied, and a situation that the user semantics cannot be understood may occur, so that in order to ensure that the result more conforming to the user semantics can be queried, an RDF directed graph needs to be converted into an RDF directed graph. The current RDF undirected graph has lower query efficiency under the condition of mass data.
Disclosure of Invention
The invention aims to: the technical problem to be solved by the invention is to provide a site retrieval method, equipment and a computer storage medium based on a knowledge graph, which are used for establishing an efficient index structure according to space and semantic information and improving the query efficiency under the condition that the query efficiency is lower under the condition that the conventional KSP method is used for carrying out massive data of a knowledge graph RDF undirected graph.
The technical scheme is as follows: the site retrieval method based on the knowledge graph comprises an index establishment method and a query method,
the indexing method comprises the following steps:
(1) Performing breadth-first search by taking a place entity as a root in the RDF graph, and storing the shortest distance between the place entity and other nodes as a semantic distance index;
(2) Iteratively dividing the whole data space into a plurality of areas according to the spatial characteristics, encoding the areas to construct a linear quadtree index, and storing semantic distance index information of place entities in the corresponding areas in leaf nodes of the quadtree;
the query method comprises the following steps:
(1) Inputting a keyword set and user position coordinates by a user;
(2) Finding out an entity node set corresponding to the keyword set in the RDF graph;
(3) Inquiring the linear quadtree index from top to bottom according to the user position coordinates to obtain leaf nodes;
(4) Inquiring and calculating the semantic distance between the regional place entity and the entity node in the semantic distance index information stored in the leaf node according to the entity node set in the step (2);
(5) Calculating the space distance between the user position coordinates and the space coordinates of the currently queried place entity;
(6) Calculating a score according to the semantic distance and the spatial distance;
(7) Querying the linear quadtree index from bottom to top, calculating the semantic distance of the place entity according to the semantic distance index information on the node, and repeating the steps (5) and (6) to obtain the score of each place entity until the root node of the linear quadtree index is queried;
(8) And sequencing the queried results, and returning the plurality of place entities with the highest scores to the user according to the high-low score sequence.
Further, the coding method in the indexing method step (2) is Morton coding based on split sequences.
Further, the specific steps of the query method step (2) are as follows: and establishing an index by taking the keywords as keys, storing the entity node set corresponding to each keyword, and mapping the entity node set to be queried through the index after the user queries the input keywords.
Further, the calculation score in the step (6) of the query method is specifically: a score is calculated based on the cost function model,
score = β x semantic distance + (1- β) x spatial distance, 0< β <1.
Further, β=0.5.
The apparatus of the present invention comprises a computer memory and a processor, the memory having stored therein computer readable instructions which, when executed by the processor, cause the processor to perform the method of any of the above.
The storage medium of the present invention has stored thereon a computer program which, when executed by a computer processor, implements a method as described in any of the above.
The beneficial effects are that: compared with the existing site query method based on the knowledge graph, the method has the remarkable advantages that: in the existing method, the space position factor of the user is not considered, in addition, the traditional KSP only considers the condition of the RDF directed graph, the real user semantics cannot be met, an effective index is not constructed, and a better efficient query method for the RDF directed graph is not available. Therefore, for the site query of the large-scale RDF undirected graph, the invention respectively establishes the space and the semantic indexes and combines and stores the space and the semantic indexes under the consideration of the influence of two factors of space and semantic on the query result, designs a linear quadtree index structure combining the space and the semantic information, designs a high-efficiency query algorithm aiming at the index structure, can gradually expand the retrieval range from bottom to top according to the space index information, rapidly calculates whether the current result meets the actual semantic requirement of a user according to the semantic information, and improves the efficiency of the on-line site query while ensuring the accuracy of the query result. Experiments prove that the method can greatly improve the efficiency of inquiring the knowledge graph on-line places.
Drawings
FIG. 1 is a schematic diagram of an RDF undirected graph and an RDF directed graph;
FIG. 2 is a query schematic diagram of an embodiment;
FIG. 3 is an RDF undirected graph of an embodiment query;
FIG. 4 is an overall flow chart of the present method;
FIG. 5 is a schematic diagram of a linear quadtree index construction process;
FIG. 6 is a schematic diagram of storing a spatial distance index in combination with a semantic distance index;
FIG. 7 is a schematic diagram of a linear quadtree query;
FIG. 8 is a diagram of query time versus time.
Detailed Description
The experimental hardware environment is a high-performance computing platform DELL cluster of a laboratory, and comprises 11 nodes, and each node is configured as a CPU of 2.7GHz 8 core and a 64GB memory. The data set used was the Yago open source data set.
Analysis is performed herein by a query example, assuming that the user is located in boston and expects to query "university of the nurturing nobel prize acquirer and the drawing prize acquirer", and user input is a keyword set "university of the nobel prize acquirer drawing prize acquirer", the query example is shown in fig. 2, and the RDF undirected graph of the query is shown in fig. 3.
Based on the traditional KSP method, the input of the system is a user query keyword set and the geographic position of the user, and the input is a place result set meeting the user query requirement. Firstly, the system finds all university place entities, carries out breadth-first search on each entity, finds the university of the Tuber prize acquirer of the Nobel prize acquirer of the user input keyword set, calculates the sum of semantic distances between the place entities and the keyword set, calculates the space distance between the place entities and the user query position, and obtains final Top-K place entity result sets through comprehensive sequencing, wherein Top-K is the best K results.
The method of the embodiment of the invention is shown in fig. 4, and comprises an index establishing method and a query method.
The index establishing method comprises the following specific steps:
(1) Semantic distance index
Shortest distance d between any two points in RDF (remote data flow) chart G (s, t). Let g= (V, E) be a graph comprising vertex set V and edge set E, distance d between any two nodes G (s, t) satisfies the following triangle inequality:
the breadth first search is performed with the place entity as the root, and the shortest distance between the place entity and other entity nodes is stored in the form of a key value pair, which is called a mark D (v).
The label D (v). For each location entity v, the label D (v) is a node pair (u, delta) uv ) A set, where u is any node, delta uv Representing the shortest distance d from node u to node v G (u, v). The shortest distance between the QUERY node s and the node t, denoted by QUERY (s, t, D (v)), is as follows:
QUERY(s,t,D(v))＝min{δ vs +δ vt |(v,δ vs )∈D(s),(v,δ vt )}∈D(t)}(2)
for any node v, stored as a token D (v), the QUERY QUERY (s, t, L) has a complexity of O (|D(s) |+|D (t) |). Table 1 is a partial semantic distance index example.
Table 1 semantic distance index example
(2) Spatial distance index
Minimum outsourcing rectangle MBR. Assume that there are n regions in space, each of which is region I i The minimum bounding rectangle MBR of (1) is expressed in terms of its upper left and lower right corner coordinates:
{I i |I i the upper left corner point is
I i The lower right corner is +>
} (3)
According to I i Spatial coordinates of region MBR and location q (x 1 ,y 1 ) Can judge whether the place is in the region I i Inner:
firstly, finding MBR containing all location entity nodes in RDF graph, i.e. finding a minimum rectangular MBR can contain all location entities, using this as I of root node of linear quadtree, at the same time using it as whole data space, according to distribution condition of nodes using vertical or parallel of x-axis direction or y-axis direction to divide data space into 4 portions so as to make each portion contain severalThe equal number of location entity nodes can make the height of the quadtree lower when constructing the quadtree so as to improve the performance of parallel query, respectively calculate MBRs of four sub-areas for generating the sub-quadtree and establishing the link connection with the father node, and continuously and iteratively execute until the number of the location entity nodes in each area reaches the threshold value theta t Up to a threshold value theta t And determining according to the number of nodes contained in the minimum area and the storage size.
The Morton coding method comprises the steps of carrying out Morton coding on the quadtree based on a splitting sequence, wherein the coding is carried out according to the splitting sequence: SW, SE, NW, NE, which represent 00, 01, 10 and 11 respectively, links the split sequences to obtain the final code, e.g. the split sequence is "SW-SE", the code corresponds to "0001". The node paths in the quadtree and the specific region codes are globally unique, and meanwhile, the node numbers of each sub-region are the same as much as possible, so that the height of the whole tree can be reduced, morton coding bits are also reduced, storage space is reduced, quick query is convenient, and after the linear quadtree is constructed, the data structure of the B+ tree is stored in a key value pair form of < code ID and MBR >. The coding rule of the linear quadtree is as follows, and as shown in fig. 5, (b) in the figure represents the location nodes and distribution situation existing in one data space. In the figure, (c) represents a linear quadtree index with built-up, the circle represents a non-leaf node, the square represents a leaf node, the depth of the tree in the figure is 2, the splitting sequence of the node 3 is "SW-NE", so that the code is "0011", and (d) represents a mapping storage diagram of region code and region corresponding MBR.
After the linear quadtree index is constructed, the semantic distance index information of the local entity in the area is stored in the leaf node, and the storage structure of the spatial distance index and the semantic distance index is shown in fig. 6.
The specific steps of the query process method are as follows:
a. the input of the system query is a keyword set input by a user and a space position coordinate where the user is located; b. the system firstly finds an entity node set corresponding to the keyword set by utilizing constructed inverted index information according to the keyword set input by a user, and the entity node set to be queried by the system is obtained, and the concrete method of the inverted index is as follows: the Key words are used as keys, an index is established to store a node set which corresponds to each Key word, and the Key words and the entity set are mapped into the Key words and the entity set, so that the object of constructing the inverted index is that the entity set to be queried can be mapped out quickly when a user queries the input Key words, and the semantic distance is further obtained through quick calculation of the constructed semantic distance index; c. according to the position coordinate information of the user, searching MBR of each layer of nodes from top to bottom in the linear quadtree index, comparing with the position space coordinate, and positioning to obtain the minimum area of the user position, namely a certain leaf node of the linear quadtree; d. according to a semantic distance index table maintained by the corresponding leaf node of the area of the linear quadtree and according to the node set to be queried obtained in the step b, rapidly querying in the semantic distance index table and calculating to obtain the sum of the shortest distances from each place entity in the area to the node set to be queried, namely obtaining the semantic distance between the place entity and the query requirement of the user; e. obtaining a spatial distance between the user position coordinate and the spatial coordinate of the currently queried location entity through a Euclidean distance calculation formula; f. and calculating according to the cost function model to obtain the score of the comprehensive space and the semantic information, wherein the ordering model is as follows: f (semantic distance, spatial distance) =β×semantic distance + (1- β) x spatial distance, and the value of β is determined according to the experimental effect, and defaults to 0.5. Obtaining the coincidence degree of the current query location entity and the actual demand of the user through the cost function model; g. as shown in fig. 7, according to the linear quadtree index, different regions are queried from bottom to top, the semantic distance of each place entity in the region is calculated according to the semantic distance index table in the region, the step e and the step f are repeated to obtain the score of each place entity until the whole region space is queried, namely, the root node of the linear quadtree is queried; h. the system sorts the queried results, and sequentially returns top-k place entities with highest scores to the user according to the sorting of the scores, namely, the places are the results which are most in line with the query requirements of the user under the consideration of space and semantic factors.
In this embodiment, the index-established query time and the conventional index-free query time pair are shown in fig. 8, and experiments prove that the method can greatly improve the efficiency of the online site query of the knowledge graph.
Embodiments of the present invention also provide an apparatus comprising a memory and at least one processor, a computer program stored in the memory and executable on the at least one processor, at least one communication bus. The steps of the above embodiments are implemented when the at least one processor executes the computer program.
Embodiments of the present invention also provide a computer storage medium having a computer program stored thereon. The method of the above-described embodiments may be implemented when the computer program is executed by a processor. The computer storage medium is, for example, a computer-readable storage medium.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Claims (7)
1. A place retrieval method based on a knowledge graph is characterized in that: including the establishment of an indexing method and a query method,
the indexing method comprises the following steps:
(1) Performing breadth-first search by taking a place entity as a root in the RDF graph, and storing the shortest distance between the place entity and other nodes as a semantic distance index;
the establishment of the semantic distance index specifically comprises the following steps: the shortest distance between any two points in the RDF graph is marked as d G (s, t), assume g= (V, E) is a graph comprising vertex set V and edge set E, distance d between any two nodes G (s, t) satisfies the following triangle inequality:
wherein v represents a location entity; s and t are any entity node;
taking a place entity as a root to conduct breadth first search, and storing the shortest distance between the place entity and other entity nodes as a key value pair, which is called a mark D (v); i.e. for each location entity v, the label D (v) is a node pair (u, delta) uv ) Collection setAnd (d) wherein u is any physical node, delta uv Representing the shortest distance d from node u to node v G (u,v)；
(2) Iteratively dividing the whole data space into a plurality of areas according to the spatial characteristics, encoding the areas to construct a linear quadtree index, and storing semantic distance index information of place entities in the corresponding areas in leaf nodes of the quadtree;
the query method comprises the following steps:
(1) Inputting a keyword set and user position coordinates by a user;
(2) Finding out an entity node set corresponding to the keyword set in the RDF graph;
(3) Inquiring the linear quadtree index from top to bottom according to the user position coordinates to obtain leaf nodes;
(4) Inquiring and calculating the semantic distance between the regional place entity and the entity node in the semantic distance index information stored in the leaf node according to the entity node set in the step (2);
(5) Calculating the space distance between the user position coordinates and the space coordinates of the currently queried place entity;
(6) Calculating a score according to the semantic distance and the spatial distance;
(7) Querying the linear quadtree index from bottom to top, calculating the semantic distance of the place entity according to the semantic distance index information on the node, and repeating the steps (5) and (6) to obtain the score of each place entity until the root node of the linear quadtree index is queried;
(8) And sequencing the queried results, and returning the plurality of place entities with the highest scores to the user according to the high-low score sequence.
2. The knowledge-based location retrieval method as claimed in claim 1, wherein: the coding method in the indexing method establishing step (2) is Morton coding based on split sequences.
3. The knowledge-based location retrieval method as claimed in claim 1, wherein: the specific steps of the query method step (2) are as follows: and establishing an index by taking the keywords as keys, storing the entity node set corresponding to each keyword, and mapping the entity node set to be queried through the index after the user queries the input keywords.
4. The knowledge-based location retrieval method as claimed in claim 1, wherein: the calculation score in the step (6) of the query method is specifically: a score is calculated based on the cost function model,
score = β x semantic distance + (1- β) x spatial distance, 0< β <1.
5. The knowledge-based location retrieval method as claimed in claim 4, wherein: beta=0.5.
6. A computer device comprising a computer memory and a processor, the memory having stored therein computer readable instructions which, when executed by the processor, cause the processor to perform the method of any of claims 1 to 5.
7. A storage medium having a computer program stored thereon, characterized by: the computer program implementing the method according to any of claims 1 to 5 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910333290.4A CN110059264B (en) | 2019-04-24 | 2019-04-24 | Site retrieval method, equipment and computer storage medium based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910333290.4A CN110059264B (en) | 2019-04-24 | 2019-04-24 | Site retrieval method, equipment and computer storage medium based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110059264A CN110059264A (en) | 2019-07-26 |
| CN110059264B true CN110059264B (en) | 2023-07-07 |
Family
ID=67320505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910333290.4A Active CN110059264B (en) | 2019-04-24 | 2019-04-24 | Site retrieval method, equipment and computer storage medium based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110059264B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110688492B (en) * | 2019-09-25 | 2020-12-08 | 东南大学 | Knowledge graph query method based on lightweight index |
| CN111130569B (en) * | 2019-12-17 | 2021-11-30 | 佛山科学技术学院 | Spatial information data self-adaptive fault-tolerant processing method and system |
| CN111221864B (en) * | 2019-12-27 | 2022-07-08 | 江苏艾佳家居用品有限公司 | Intelligent index recommendation method based on mysql slow query log word frequency analysis |
| CN111625611A (en) * | 2020-05-25 | 2020-09-04 | Oppo广东移动通信有限公司 | Task processing and map construction method and device, equipment and medium for map application |
| CN113407669B (en) * | 2021-06-18 | 2022-11-11 | 北京理工大学 | Semantic track query method based on activity influence |
| CN113626612A (en) * | 2021-08-13 | 2021-11-09 | 第四范式（北京）技术有限公司 | Prediction method and system based on knowledge graph reasoning |
| CN113836445A (en) * | 2021-09-16 | 2021-12-24 | 北京百度网讯科技有限公司 | Semantization method and device, electronic equipment and readable storage medium |
| CN117290617B (en) * | 2023-08-18 | 2024-05-10 | 中国船舶集团有限公司第七〇九研究所 | Offshore distributed multi-source heterogeneous space-time data query method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105404685B (en) * | 2015-12-09 | 2020-07-28 | 中国农业银行股份有限公司 | Electronic map space query method and system |
| CN106021457B (en) * | 2016-05-17 | 2019-10-15 | 福州大学 | RDF distributed semantic searching method based on keyword |
| CN106503223B (en) * | 2016-11-04 | 2019-12-17 | 华东师范大学 | online house source searching method and device combining position and keyword information |
| CN108733803B (en) * | 2018-05-18 | 2022-04-29 | 电子科技大学 | Multi-user space keyword query method under road network |
-
2019
- 2019-04-24 CN CN201910333290.4A patent/CN110059264B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110059264A (en) | 2019-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110059264B (en) | Site retrieval method, equipment and computer storage medium based on knowledge graph | |
| CN108932347B (en) | Spatial keyword query method based on social perception in distributed environment | |
| US9934324B2 (en) | Index structure to accelerate graph traversal | |
| CN106933833B (en) | Method for quickly querying position information based on spatial index technology | |
| CN104376112B (en) | A kind of method of road cyberspace key search | |
| Huang et al. | Continuous distance-based skyline queries in road networks | |
| CN106874426B (en) | RDF (resource description framework) streaming data keyword real-time searching method based on Storm | |
| CN107145526B (en) | Reverse-nearest neighbor query processing method for geographic social keywords under road network | |
| CN104794221A (en) | Multi-dimensional data analyzing system based on service objects | |
| CN110069500B (en) | Dynamic mixed indexing method for non-relational database | |
| CN109992590B (en) | Approximate space keyword query method and system with digital attributes in traffic network | |
| CN109190052B (en) | Spatial indexing method based on social perception in distributed environment | |
| CN116244333A (en) | Database query performance prediction method and system based on cost factor calibration | |
| Jin et al. | Collective keyword query on a spatial knowledge base | |
| Cho et al. | A basis of spatial big data analysis with map-matching system | |
| Cho et al. | A GPS trajectory map-matching mechanism with DTG big data on the HBase system | |
| CN110263108B (en) | Keyword Skyline fuzzy query method and system based on road network | |
| Sudhir et al. | Replicated layout for in-memory database systems | |
| Chan et al. | Dynamic planar orthogonal point location in sublogarithmic time | |
| CN116028678A (en) | Method and system for searching full-quantity path in knowledge graph | |
| John et al. | Dynamic sorting and average skyline method for query processing in spatial-temporal data | |
| Goncalves et al. | Making recommendations using location-based skyline queries | |
| Cai et al. | Continuous road network-based skyline query for moving objects | |
| Huang et al. | Processing continuous K-nearest skyline query with uncertainty in spatio-temporal databases | |
| CN114757147A (en) | BERT-based automatic hierarchical tree expansion method |
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 |