CN107145526A - Geographical social activity keyword Reverse nearest neighbor inquiry processing method under a kind of road network - Google Patents
Geographical social activity keyword Reverse nearest neighbor inquiry processing method under a kind of road network Download PDFInfo
- Publication number
- CN107145526A CN107145526A CN201710244072.4A CN201710244072A CN107145526A CN 107145526 A CN107145526 A CN 107145526A CN 201710244072 A CN201710244072 A CN 201710244072A CN 107145526 A CN107145526 A CN 107145526A
- Authority
- CN
- China
- Prior art keywords
- user
- point
- road network
- count table
- similitude
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses geographical social activity keyword Reverse nearest neighbor inquiry processing method under a kind of road network, using GIM trees to space road network, text, social data is stored, and travels through index using branch-bound method;The present invention minimum similitude count table of computation index node first and maximum comparability count table when traveling through index, then beta pruning is carried out using above-mentioned minimum similitude count table and maximum comparability count table, and using filtering, refining algorithm to accelerate query execution.Present invention incorporates the prior art of spatial database, geographical social text similarity calculation times are reduced, so as to improve query performance.
Description
Technical field
It is a kind of geographical social crucial under road network for handling the present invention relates to the index of spatial database and inquiring technology
The method of word Reverse nearest neighbor inquiry.
Background technology
Spatial data refer to that GIS-Geographic Information System stores on computer physical storage medium to applying related geography
The summation of spatial data, its purpose is to store, the various geographical spatial datas of management and retrieval.Wherein, road network spatial data
As the important component of spatial database, increasing concern has been obtained.In order to quickly and efficiently access road network space
Data, experts and scholars propose many road network space data index methods.At present, G trees indexing means are maximally effective road networks
Space data index method.Road network is divided into multiple subgraphs by it, and precalculates the road network distance of each boundary point, so as to reach
Reduce the purpose of shortest path calculation cost.
Reverse nearest neighbor inquiry due to its decision support and find potential user in terms of important application and receive
The extensive concern of art circle.In the correlative study that Reverse nearest neighbor is inquired about, the inquiry of spatial key Reverse nearest neighbor is by people under road network
For finding interest collection.Wherein, interest collection refers to group interested in some point of interest.However, spatial key under road network
The inquiry of word Reverse nearest neighbor only considered text and spatial information, and search those most possible crowds as potential user.
With the development of social networks, the scale of construction of social network data is increasing.In social networks, there are social connections
User may have similar hobby, thus this kind of data can for prediction and recommendation support is provided.Based on this, people
It has studied geographical social keyword query.A given geographical social keyword query and the user for submitting the inquiry, this inquiry
Return to space length recently, text similarity highest point of interest, and the friend of the user accesses the number of times of the point of interest most
It is many.
At present, for spatial key Reverse nearest neighbor inquiry under road network and the geographical existing ripe solution of social keyword query
Certainly scheme.But in some application scenarios, Reverse nearest neighbor inquiry will not only consider space and text message, and consider
The information of registering of social information and user between user to point of interest.However, existing inquiry processing method can't have
Effect ground solves the problems, such as above-mentioned inquiry.
The content of the invention
Geographical social activity keyword Reverse nearest neighbor inquiry under road network can not be effectively handled instant invention overcomes prior art to ask
There is provided geographical social activity keyword Reverse nearest neighbor inquiry processing method under a kind of road network for topic.
The technical solution adopted for the present invention to solve the technical problems step is as follows:Geographical social activity keyword under a kind of road network
Reverse nearest neighbor inquiry processing method, this method comprises the following steps:
Step (1):User and point of interest are collected, GIM tree index structures are built to it;
Step (2):Calculate the minimum similitude count table of the social keyword of geography of the node of each GIM trees index structure
With maximum comparability count table;
Step (3):The user being collected into using pruning algorithms to step (1) is filtered with point of interest;
Step (4):According to the result filtered in step (3), undesirable user is rejected by refining algorithm, with
To final result set.
Further, the construction step of GIM tree index structures is as follows in described step (1):Whole road network is divided into
Multiple subgraphs, and the road-net node for belonging to multiple subgraphs is defined as boundary point;Precalculate the road network between all boundary points
Distance;Each GIM tree index structures node is handed over and inverted file and two matrixes comprising a Road Network Sub-graph, one;Hand over and fall
Row's file describes the text message between user and point of interest;Two matrixes are that user registers matrix and user social contact relation
Matrix, user registers matrix storage user to the number of times of registering of each point of interest, between user social contact relational matrix storage user
Social networks.
Further, in described step (2) minimum similitude count table and maximum comparability count table computational methods
It is as follows:
One group of user and one group of point of interest are given, is registered matrix and user social contact relational matrix using user in step (1)
The two matrix multiples calculate the minimum value and maximum of geographical social keyword similitude between user and point of interest;Using upper
State minimum value and maximum builds the minimum similitude count table and maximum comparability count table of user.
Further, pruning algorithms are specific as follows in described step (3):
Give a query point, according to the computational methods of step (2), obtain query point and user's similitude minimum value and
Maximum, the minimum similitude count table and maximum comparability count table obtained in conjunction with step (2) carries out beta pruning to user, its
In:
1) if query point and the maximum that user gathers similitude are smaller than the floor value of minimum similitude count table, abandon
This group of user.
If 2) query point and the minimum value that user gathers similitude are bigger than the upper dividing value of maximum comparability count table, by this group
User is inserted into final result set.
Further, the filter process in described step (3) is as follows:
1) Subscriber Queue and a point of interest queue are initialized, GIM trees are indexed to the subscriber data set of root node
It is put into Subscriber Queue, interest point data set is put into point of interest queue;
2) candidate user set and a final result set are initialized, and preserves the GIM that current accessed is crossed respectively
Set in index node not by the user of beta pruning and the user for being confirmed to be final result;
3) if Subscriber Queue is sky, candidate user set and final result set are returned;Otherwise the of Subscriber Queue is taken out
One element, and beta pruning is carried out using the pruning algorithms in step (3) to child node of the element in GIM tree index structures,
If condition can be met, then insert it into final result set;If not by beta pruning, inserting it into candidate user collection
Close.
Further, the refining algorithm in described step (4) is comprised the following steps that:
1) each user for taking out candidate user set in step (3);
2) geographical social activity keyword search results set under the road network of the user is found out with space length order;
3) if query point is in the above results set, the user is inserted into final result set;Otherwise the use is abandoned
Family;
4) final result set is returned.
The invention has the advantages that:The present invention takes full advantage of existing index technology in spatial database, instead
K-NN search and spatial key inquiring technology, multiple subnets are divided into by road network, and are precalculated most short between subnet
Path distance, so as to reduce shortest path calculation cost;The index structure of minimum and maximum count table is devised, and subnet is entered
Row beta pruning;Efficient trimming algorithm is devised, so as to greatly reduce I/O number and CPU calculating times;Propose to utilize matrix meter
The method for calculating social similitude, reduces calculation cost;A kind of branch-bound algorithm is proposed, so as to avoid to index structure
Repeated accesses, improve the efficiency of inquiry.
Brief description of the drawings
Fig. 1 is the implementation steps flow chart of the present invention.
Embodiment
Technical scheme is described further in conjunction with accompanying drawing and specific implementation:
As shown in figure 1, specific implementation process of the present invention and operation principle are as follows:
Step (1):User and point of interest are collected, GIM tree index structures are built to it;
Step (2):Calculate the minimum similitude count table of the social keyword of geography of the node of each GIM trees index structure
With maximum comparability count table;
Step (3):The user being collected into using pruning algorithms to step (1) is filtered with point of interest;
Step (4):According to the result filtered in step (3), undesirable user is rejected by refining algorithm, with
To final result set.
Further, the information of each point of interest includes positional information, text message and letter of registering in the step (1)
Breath, wherein positional information is a geographical coordinate, and text message is a set of keyword, and information of registering is one group of record, every note
Record includes when certain user reached the point of interest;User profile includes positional information, text message and social information, wherein
Positional information is the current location of user, and text message is a set of keyword, and social information is the friends between user.It is all
Information is all stored in GIM tree index structures.The construction step of GIM tree index structures is as follows:Whole road network is divided into multiple
Subgraph, and the road-net node for belonging to multiple subgraphs is defined as boundary point;The road network distance between all boundary points is precalculated,
To accelerate the calculating of shortest path distance;Each GIM tree index structures node is handed over and the row's of falling text comprising a Road Network Sub-graph, one
Part and two matrixes;Hand over and inverted file describes text message between user and point of interest;Two matrixes are signed for user
To matrix and user social contact relational matrix, user is registered register number of times of the matrix storage user to each point of interest, and user social contact is closed
It is the social networks between matrix storage user.
Further, the computational methods of minimum similitude count table and maximum comparability count table are such as in the step (2)
Under:Give one group of user and one group of point of interest, using the user in step (1) register matrix and user social contact relational matrix this two
Individual matrix multiple calculates the minimum value and maximum of geographical social keyword similitude between user and point of interest;Using it is above-mentioned most
Small value and maximum build the minimum similitude count table and maximum comparability count table of user.Wherein, in order to improve it is social away from
From the speed of calculating, the present invention proposes a kind of computational methods based on matrix, this method using user social contact relational matrix and
User registers matrix multiple to obtain the social similitude between one group of user and one group of point of interest.
For example:Give 2 GIM tree nodes N1With N2, node N is taken out respectively1In user's set U1With node N2In it is emerging
Interesting point set O2;U is calculated respectively1With O2Text similarity, the minimum value and maximum of spatial simlanty and social similitude
Value;For user's set U1, utilize U1Minimum similitude count table, count table are built with the minimum value of interest point set similitude
In each element include:One group of point of interest Oi, OiThe number of middle point of interest | Oi|, and U1With OiMinimum similarity;
Similarly, U is utilized1Each in maximum comparability count table, count table is built with the maximum of interest point set similitude
Element is included:One group of point of interest Oi, OiThe number of middle point of interest | Oi|, and U1With OiMaximum comparability value.
Further, pruning algorithms are specific as follows in step (3):
Give a query point, according to the computational methods of step (2), obtain query point and user's similitude minimum value and
Maximum, the minimum similitude count table and maximum comparability count table obtained in conjunction with step (2) carries out beta pruning to user, its
In:
1) if query point and the maximum that user gathers similitude are smaller than the floor value of minimum similitude count table, abandon
This group of user.
If 2) query point and the minimum value that user gathers similitude are bigger than the upper dividing value of maximum comparability count table, by this group
User is inserted into final result set.
Further, the filter process in described step (3) is as follows:
1) Subscriber Queue and a point of interest queue are initialized, GIM trees are indexed to the subscriber data set of root node
It is put into Subscriber Queue, interest point data set is put into point of interest queue;
2) candidate user set and a final result set are initialized, and preserves the GIM that current accessed is crossed respectively
Set in index node not by the user of beta pruning and the user for being confirmed to be final result;
3) if Subscriber Queue is sky, candidate user set and final result set are returned;Otherwise the of Subscriber Queue is taken out
One element, and beta pruning is carried out using the pruning algorithms in step (3) to child node of the element in GIM tree index structures,
If condition can be met, then insert it into final result set;If not by beta pruning, inserting it into candidate user collection
Close.
Further, the refining algorithm in described step (4) is comprised the following steps that:
1) each user for taking out candidate user set in step (3);
2) geographical social activity keyword search results set under the road network of the user is found out with space length order;
3) if query point is in the above results set, the user is inserted into final result set;Otherwise the use is abandoned
Family;
4) final result set is returned.
Claims (6)
1. geographical social activity keyword Reverse nearest neighbor inquiry processing method under a kind of road network, it is characterised in that:This method includes as follows
Step:
Step (1):User and point of interest are collected, GIM tree index structures are built to it.
Step (2):Calculate the node of each GIM trees index structure the social keyword of geography minimum similitude count table with most
Big similitude count table.
Step (3):The user being collected into using pruning algorithms to step (1) is filtered with point of interest.
Step (4):According to the result filtered in step (3), undesirable user is rejected by refining algorithm, to obtain most
Whole results set.
2. geographical social activity keyword Reverse nearest neighbor inquiry processing method under road network according to claim 1, it is characterised in that:
The construction step of GIM tree index structures is as follows in described step (1):Whole road network is divided into multiple subgraphs, and will be belonged to
The road-net node of multiple subgraphs is defined as boundary point;Precalculate the road network distance between all boundary points;Each GIM trees index
Structure node is handed over and inverted file and two matrixes comprising a Road Network Sub-graph, one;Hand over and inverted file describes user
Text message between point of interest;Two matrixes are that user registers matrix and user social contact relational matrix, and user registers matrix
Store the social networks between number of times of registering of the user to each point of interest, user social contact relational matrix storage user.
3. geographical social activity keyword Reverse nearest neighbor Query Processing Algorithm under road network according to claim 2, it is characterised in that:
The computational methods of minimum similitude count table and maximum comparability count table are as follows in described step (2):
One group of user and one group of point of interest are given, is registered matrix and user social contact relational matrix the two matrix multiples using user
Calculate the minimum value and maximum of geographical social keyword similitude between user and point of interest;Utilize above-mentioned minimum value and maximum
Value builds the minimum similitude count table and maximum comparability count table of user.
4. geographical social activity keyword Reverse nearest neighbor inquiry processing method under road network according to claim 3, it is characterised in that:
Pruning algorithms are specific as follows in described step (3):
A query point is given, according to the computational methods of step (2), the minimum value and maximum of query point and user's similitude is obtained
Value, the minimum similitude count table and maximum comparability count table obtained in conjunction with step (2) carries out beta pruning to user, wherein:
1) if query point and the maximum that user gathers similitude are smaller than the floor value of minimum similitude count table, this group is abandoned
User.
If 2) query point and the minimum value that user gathers similitude are bigger than the upper dividing value of maximum comparability count table, by this group of user
It is inserted into final result set.
5. geographical social activity keyword Reverse nearest neighbor inquiry processing method under road network according to claim 4, it is characterised in that:
Filter process in described step (3) is as follows:
1) Subscriber Queue and a point of interest queue are initialized, use is put into user's set that GIM trees are indexed into root node
In the queue of family, interest point set is put into point of interest queue;
2) candidate user set and a final result set are initialized, and preserves the GIM tree ropes that current accessed is crossed respectively
Draw in node not by the user of beta pruning and the user for being confirmed to be final result;
3) if Subscriber Queue is sky, candidate user set and final result set are returned;Otherwise take out Subscriber Queue first
Element, and to the element GIM tree index structures child node using in step (3) pruning algorithms carry out beta pruning, if energy
Meet condition, then insert it into final result set;If not by beta pruning, inserting it into candidate user set.
6. geographical social activity keyword Reverse nearest neighbor inquiry processing method under road network according to claim 5, it is characterised in that:
Refining algorithm in described step (4) is comprised the following steps that:
1) each user for taking out candidate user set in step (3);
2) geographical social activity keyword search results set under the road network of the user is found out with space length order;
3) if query point is in the above results set, the user is inserted into final result set;Otherwise the user is abandoned;
4) final result set is returned.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710244072.4A CN107145526B (en) | 2017-04-14 | 2017-04-14 | Reverse-nearest neighbor query processing method for geographic social keywords under road network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710244072.4A CN107145526B (en) | 2017-04-14 | 2017-04-14 | Reverse-nearest neighbor query processing method for geographic social keywords under road network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107145526A true CN107145526A (en) | 2017-09-08 |
CN107145526B CN107145526B (en) | 2020-06-05 |
Family
ID=59774821
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710244072.4A Active CN107145526B (en) | 2017-04-14 | 2017-04-14 | Reverse-nearest neighbor query processing method for geographic social keywords under road network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107145526B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107908722A (en) * | 2017-11-14 | 2018-04-13 | 华东师范大学 | Reverse k rankings querying method based on distance |
CN108733803A (en) * | 2018-05-18 | 2018-11-02 | 电子科技大学 | A kind of Multi-User Dimension keyword query method under road network |
CN109408738A (en) * | 2018-09-10 | 2019-03-01 | 中南民族大学 | The querying method and system of spatial entities in a kind of transportation network |
CN111813778A (en) * | 2020-07-08 | 2020-10-23 | 安徽工业大学 | Approximate keyword storage and query method for large-scale road network data |
CN112883272A (en) * | 2021-03-16 | 2021-06-01 | 山东大学 | Method for determining recommended object |
CN113868549A (en) * | 2021-09-22 | 2021-12-31 | 浙江大学 | Advertisement putting optimization method and device, electronic equipment and storage medium |
CN114780875A (en) * | 2022-06-22 | 2022-07-22 | 广东省智能机器人研究院 | Dynamic group travel planning query method |
CN117076726A (en) * | 2023-09-14 | 2023-11-17 | 上海交通大学 | Approximate neighbor searching method, system, medium and device based on ray tracing intersection |
CN117076726B (en) * | 2023-09-14 | 2024-06-07 | 上海交通大学 | Approximate neighbor searching method, system, medium and device based on ray tracing intersection |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408117A (en) * | 2014-11-26 | 2015-03-11 | 浙江大学 | Best consumer real-time searching method based on road network continuous aggregation nearest neighbor query |
CN103345509B (en) * | 2013-07-04 | 2016-08-10 | 上海交通大学 | Obtain the level partition tree method and system of the most farthest multiple neighbours on road network |
-
2017
- 2017-04-14 CN CN201710244072.4A patent/CN107145526B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103345509B (en) * | 2013-07-04 | 2016-08-10 | 上海交通大学 | Obtain the level partition tree method and system of the most farthest multiple neighbours on road network |
CN104408117A (en) * | 2014-11-26 | 2015-03-11 | 浙江大学 | Best consumer real-time searching method based on road network continuous aggregation nearest neighbor query |
Non-Patent Citations (2)
Title |
---|
JINLING JIANG 等: "Finding Top-k Local Users in Geo-Tagged Social", 《2015 IEEE 31ST INTERNATIONAL CONFERENCE ON DATA ENGINEERING》 * |
YUNJUN GAO 等: "Visible Reverse k-Nearest Neighbor Query Processing in Spatial Databases", 《IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107908722B (en) * | 2017-11-14 | 2021-10-12 | 华东师范大学 | Reverse k ranking query method based on distance |
CN107908722A (en) * | 2017-11-14 | 2018-04-13 | 华东师范大学 | Reverse k rankings querying method based on distance |
CN108733803A (en) * | 2018-05-18 | 2018-11-02 | 电子科技大学 | A kind of Multi-User Dimension keyword query method under road network |
CN108733803B (en) * | 2018-05-18 | 2022-04-29 | 电子科技大学 | Multi-user space keyword query method under road network |
CN109408738A (en) * | 2018-09-10 | 2019-03-01 | 中南民族大学 | The querying method and system of spatial entities in a kind of transportation network |
CN111813778B (en) * | 2020-07-08 | 2024-03-29 | 安徽工业大学 | Approximate keyword storage and query method for large-scale road network data |
CN111813778A (en) * | 2020-07-08 | 2020-10-23 | 安徽工业大学 | Approximate keyword storage and query method for large-scale road network data |
CN112883272A (en) * | 2021-03-16 | 2021-06-01 | 山东大学 | Method for determining recommended object |
CN112883272B (en) * | 2021-03-16 | 2022-04-29 | 山东大学 | Method for determining recommended object |
CN113868549A (en) * | 2021-09-22 | 2021-12-31 | 浙江大学 | Advertisement putting optimization method and device, electronic equipment and storage medium |
CN113868549B (en) * | 2021-09-22 | 2024-05-17 | 浙江大学 | Advertisement putting optimization method and device, electronic equipment and storage medium |
CN114780875B (en) * | 2022-06-22 | 2022-09-06 | 广东省智能机器人研究院 | Dynamic group travel planning query method |
CN114780875A (en) * | 2022-06-22 | 2022-07-22 | 广东省智能机器人研究院 | Dynamic group travel planning query method |
CN117076726A (en) * | 2023-09-14 | 2023-11-17 | 上海交通大学 | Approximate neighbor searching method, system, medium and device based on ray tracing intersection |
CN117076726B (en) * | 2023-09-14 | 2024-06-07 | 上海交通大学 | Approximate neighbor searching method, system, medium and device based on ray tracing intersection |
Also Published As
Publication number | Publication date |
---|---|
CN107145526B (en) | 2020-06-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107145526A (en) | Geographical social activity keyword Reverse nearest neighbor inquiry processing method under a kind of road network | |
CN104346444B (en) | A kind of the best site selection method based on the anti-spatial key inquiry of road network | |
CN103116639A (en) | Item recommendation method and system based on user-item bipartite model | |
CN102521364B (en) | Method for inquiring shortest path between two points on map | |
CN104834693A (en) | Depth-search-based visual image searching method and system thereof | |
CN107025228B (en) | Question recommendation method and equipment | |
CN106777163A (en) | IP address institute possession querying method and system based on RBTree | |
CN106933511B (en) | Space data storage organization method and system considering load balance and disk efficiency | |
CN107103032A (en) | The global mass data paging query method sorted is avoided under a kind of distributed environment | |
Yuan et al. | RSkNN: kNN search on road networks by incorporating social influence | |
CN107766406A (en) | A kind of track similarity join querying method searched for using time priority | |
CN104933143A (en) | Method and device for acquiring recommended object | |
CN110795613B (en) | Commodity searching method, device and system and electronic equipment | |
CN108549696B (en) | Time series data similarity query method based on memory calculation | |
CN111177410A (en) | Knowledge graph storage and similarity retrieval method based on evolution R-tree | |
CN111191147B (en) | Group discovery method and analysis system based on big data mining | |
CN112765288A (en) | Knowledge graph construction method and system and information query method and system | |
CN109033746B (en) | Protein compound identification method based on node vector | |
CN106919712A (en) | The data statistical approach and system of form | |
Wang et al. | Top-k socially constrained spatial keyword search in large siot networks | |
CN107273418A (en) | A kind of across Noumenon property chain inference method based on cloud platform | |
CN106844539A (en) | Real-time data analysis method and system | |
Lu et al. | The major research themes of big data literature: From 2001 to 2016 | |
CN114036345A (en) | Method and device for processing track data and storage medium | |
CN109150974B (en) | User identity linking method based on neighbor iteration similarity |
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 |