CN108536813B - Track querying method, electronic equipment and storage medium - Google Patents
Track querying method, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN108536813B CN108536813B CN201810303902.0A CN201810303902A CN108536813B CN 108536813 B CN108536813 B CN 108536813B CN 201810303902 A CN201810303902 A CN 201810303902A CN 108536813 B CN108536813 B CN 108536813B
- Authority
- CN
- China
- Prior art keywords
- point
- query
- track
- candidate tracks
- distance
- 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
- G06F16/20—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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of track querying method, which comprises obtains the query set of the description including each query point；The description of each query point is converted into the corresponding theme probability distribution with position and time tag of each query point；Each query point is scanned for matching, searches the candidate tracks collection of the query set by the corresponding theme probability distribution based on each query point with the semantic track data collection in database；Candidate tracks collection based on the query set, the candidate tracks for calculating the query set and the query set are concentrated at a distance from each candidate tracks, according to apart from size, concentrate each candidate tracks to be ranked up the candidate tracks of the query set；According to the candidate tracks collection after sequence, result is exported to user.The present invention also provides a kind of electronic equipment and storage mediums.The present invention can improve retrieval precision.
Description
Technical field
The present invention relates to data query field more particularly to a kind of track querying methods, electronic equipment and storage medium.
Background technique
It is different from traditional mobile object space-time trajectory (Spatio-temporal Trajectory), semantic track data
Not only contain having time, spatial information, also contain user behavior information abundant: people do it is being thought, experience, pass through
Different media means such as by modes such as text, picture, videos, constitute a series of rail of behaviors in the expression of different location
Mark.Therefore, the relevant technology in semantic track and research not only facilitate solving road congestion problems, improve out line efficiency, ensure
Traffic safety, and have important society and economic value, existing skill to the energy, optimization traffic quality and resource distribution is saved
Semantic track in art indicates retrieval of the usual user based on text similarity, focuses on the difference of " shape " between text, such as logical
Semantic track inquiry in the prior art is crossed, " drinking coffee " and tracing point description " Startbuck " are considered wide of the mark, Wu Fajian
Rope arrives, and the relevant track of such theme also can not just retrieve, and reduces query accuracy.
Summary of the invention
In view of the foregoing, it is necessary to a kind of track querying method, electronic equipment and storage medium are provided, retrieval can be improved
Precision.
A kind of track querying method, which comprises
Obtain the query set of the description including each query point；
The description of each query point is converted into the corresponding theme probability with position and time tag of each query point
Distribution；
Corresponding theme probability distribution based on each query point, by the semantic track number in each query point and database
It scans for matching according to collection, searches the candidate tracks collection of the query set；
Candidate tracks collection based on the query set, the candidate tracks for calculating the query set and the query set are concentrated often
The distance of a candidate tracks concentrates each candidate tracks to be ranked up the candidate tracks of the query set according to apart from size；
According to the candidate tracks collection after sequence, result is exported to user.
Preferred embodiment according to the present invention, it includes position coordinates and theme that the semanteme track data, which concentrates each tracing point,
Distributed intelligence includes that position coordinates and theme distribution information establish the layering including space layer and subject layer based on each tracing point
Index structure, wherein space layer establishes index structure using quaternary tree, for each leaf node indicate multiple tracing points,
Subject layer establishes the corresponding LSH index structure of each leaf node based on position sensing Hash.
Preferred embodiment according to the present invention, the corresponding theme probability distribution based on each query point, is looked into each
It askes point to scan for matching with the semantic track data collection in database, the candidate tracks collection for searching the query set includes:
Corresponding theme probability distribution based on each query point, by the semantic track number in each query point and database
It scans for matching according to collection, searches the candidate tracks of each query point, the candidate tracks of each query point are determined as described look into
Ask the candidate tracks collection of collection；
The wherein corresponding theme probability distribution based on each query point, by the semantic rail in each query point and database
Mark data set scans for matching, and the candidate tracks for inquiring a query point q include:
Based on the corresponding theme probability distribution of query point q, the leaf node of quaternary tree described in recursive traversal obtains preferential team
Column, the Priority Queues are ranked up according to the ascending order of mdist (q, N), and the mdist (q, N) indicates query point q and leaf knot
The minimum range for multiple tracing points that point N is indicated；
Each leaf node in the Priority Queues is successively traversed, traverses each leaf using multiprobe LSH index technology
The corresponding LSH index structure of child node obtains the candidate tracks point of the query point, by the candidate tracks comprising the query point
Candidate tracks of the track of point as the query point, and using the candidate tracks of the query point as the candidate of the query set
A part of track.
The calculation formula of preferred embodiment according to the present invention, the mdist (q, N) is as follows:
Mdist (q, N)=λ Ds(q,N)+(1-λ)·DT(q,N)；
Wherein DS(q, N) is the minimum boundary matrix N .rect based on leaf node N, from query point q to leaf node N
The smallest space length, DT(q, N) from q to the leaf node N the multiple tracing points indicated minimum theme distance.
Preferred embodiment according to the present invention concentrates one in the candidate tracks for calculating the query set Q and the query set Q
Candidate tracks apart from when, comprising:
Each query point is calculated at a distance from each tracing point in the candidate tracks Tr；
According to each query point at a distance from each tracing point in the candidate tracks Tr, calculate each query point with it is described
The distance of candidate tracks Tr；
According to each query point at a distance from the candidate tracks Tr, the query set and the candidate tracks Tr's are calculated
Distance.
Preferred embodiment according to the present invention, it is described to calculate each tracing point in each query point and the candidate tracks Tr
Distance includes:
A query point q containing text entry q.W and geographical location q.l is given, from a tracing point p to query point q
Distance can according between tracing point p to query point q spatial proximity and topic relativity measure, calculation formula is as follows:
D (q, p)=λ DS(q,p)+(1-λ)·DT(q, p), wherein λ ∈ [0,1] is that user specifies parameter to be used to adjust
Save the weight of spatial proximity and topic similarity；DS(q, p) is space Euclidean distance；DT(q, p) is represented between q and p text entry
Theme distance.
Preferred embodiment according to the present invention, it is described according to each query point at a distance from tracing point each in candidate tracks,
Calculate each query point includes: at a distance from the candidate tracks Tr
Given an a query point q and track Tr, one of tracing point p ∈ Tr, for other any in the track
For one tracing point p', there is d (q, p)≤d (q, p'), then tracing point p is expressed as in the track and the maximally related rail of query point q
Mark point Tr.MRP (q) is then indicated as the query point the distance between from maximally related tracing point Tr.MRP (q) to tracing point q
To the distance of track.
Preferred embodiment according to the present invention, it is described according to each query point at a distance from the candidate tracks, described in calculating
Query set includes: at a distance from the candidate tracks Tr
The given inquiry Q={ q comprising m inquiry point set1,q2,Λ,qmAnd a track Tr, Q is inquired to track Tr's
Distance DQIt (Tr) is each query point qi(i ∈ [1, m]) arrives the sum of the distance of track Tr, calculates as follows:
The most related point set MRPs of each query point is formed the most related point set Tr.MRPs (Q) of the inquiry in inquiry.
A kind of electronic equipment, the electronic equipment include memory and processor, and the memory is for storing at least one
A instruction, the processor is for executing at least one described instruction to realize that track described in any one of any embodiment is inquired
Method.
A kind of computer readable storage medium, the computer-readable recording medium storage has at least one instruction, described
At least one instruction realizes track querying method described in any one of any embodiment when being executed by processor.
As can be seen from the above technical solutions, the present invention obtains the query set of the description including each query point；Pass through base
In the similarity measurements flow function of theme distribution, by the description of each query point be converted to each query point it is corresponding with position and
The theme probability distribution of time tag；Corresponding theme probability distribution based on each query point, by each query point and data
Semantic track data collection in library scans for matching, and searches the candidate tracks collection of the query set；Based on the query set
Candidate tracks collection, the candidate tracks for calculating the query set and the query set are concentrated at a distance from each candidate tracks, according to away from
From size, the candidate tracks collection of the query set is ranked up；According to the candidate tracks collection after sequence, result is exported to use
Family.The present invention indicates that model indicates the tracing point in query point and database using semantic track, by tracing point and query point
The probability distribution that is the theme is converted in text description, i.e., a series of theme probability distribution with position and time tag makes it possible to
It enough more fully understands the intrinsic meaning of text description, and is closed by characterizing the semantic of them based on the similarity measurements of theme distribution
Connection, to improve retrieval precision.Therefore, the present invention can be inquired based on the relevant track of theme, to improve retrieval precision.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the flow chart of the preferred embodiment of track querying method of the present invention.
Fig. 2 is the functional block diagram of the preferred embodiment of track inquiry unit of the present invention.
Fig. 3 is the structural schematic diagram of the preferred embodiment of electronic equipment at least one example of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
Description and claims of this specification and term " first " in above-mentioned attached drawing, " second " and " third " etc. are
For distinguishing different objects, not for description particular order.In addition, term " includes " and their any deformations, it is intended that
Non-exclusive include in covering.Such as the process, method, system, product or equipment for containing a series of steps or units do not have
It is defined in listed step or unit, but optionally further comprising the step of not listing or unit, or optionally further comprising
For the intrinsic other step or units of these process, methods, product or equipment.
As shown in Figure 1, being the flow chart of the first preferred embodiment of track querying method of the present invention.According to different need
It asks, the sequence of step can change in the flow chart, and certain steps can be omitted.
S10, the query set Q for obtaining the description including each query point.
In alternative embodiment of the invention, the query set Q includes at least one query point.For example, user is in user
The description with multiple positions inputted on interface, needs to inquire the data of the profile matching of the multiple position, one of them
Position is described as a query point, such as the financial product of the B company under safety scientific & technical corporation, then can in the query set Q
To include two query points, query point one: description, query point two to safety scientific & technical corporation: the description to B company.
S11, the description of each query point is converted to the corresponding theme with position and time tag of each query point
Probability distribution.
Preferably, each query point is made of text description.The text description includes position and time tag.It utilizes
Semantic track indicates that model indicates the corresponding theme probability distribution with position and time tag of each query point.
Specifically, the semantic track indicates that model is as follows: each query point is one and is made of n spatial key
Text W, Z={ z1,z2,…,zmBe the theme collection, and the theme collection is for indicating theme belonging to n spatial key, then W pairs
The probability topic distribution TD answeredWThe middle each theme TD of correspondenceW[zi] calculation formula it is as follows:
Wherein Nw∈(WIWz)Indicating the number of keyword in W, α indicates symmetrical border, is usually arranged as 0.1, | W | it indicates in W
The number of keyword, | Z | indicate theme number in total.
S12, the corresponding theme probability distribution based on each query point, by the semantic rail in each query point and database
Mark data set scans for matching, and searches the candidate tracks collection of the query set Q.
Preferably, it includes position coordinates and theme distribution information that the semantic track data, which concentrates each tracing point, is based on
Each tracing point includes that position coordinates and theme distribution information establish the hierarchical index structure including space layer and subject layer.Wherein
Space layer establishes index structure using quaternary tree, to achieve the purpose that in space layer fast convergence, and in four forks of space layer
Each leaf node of tree indicates multiple tracing points, for multiple tracing points that each leaf node indicates, establishes in subject layer
The corresponding LSH index structure of each leaf node based on position sensing Hash, in this way by using hash function by each leaf
Similar tracing point is mapped to same Hash bucket in multiple tracing points that child node indicates.
Specifically, quaternary tree index is the tree construction that location information recurrence is divided into different levels.By known range
Space is divided into four equal subspaces, and such recurrence is gone down, until the level of tree reaches certain depth or meets certain
Stop segmentation after it is required that.Quaternary tree it is relatively simple for structure, and when spatial data object distribution it is relatively uniform when, have compare
High spatial data insertion and search efficiency.Tracing point is stored in leaf node, and intermediate node and root node do not store
Tracing point.
The corresponding theme probability distribution of each query point is preferably based on to scan for matching with semantic track data collection,
The candidate tracks point of each query point is obtained, using the track of the candidate tracks point comprising each query point as each query point
Candidate tracks, using the candidate tracks of each query point as the candidate tracks of the query set Q.It is based on each query point pair in this way
The theme probability distribution answered, calculates each query point and semantic track data concentrates the similarity of tracing point, is based on theme distribution
Similarity characterize query point and semantic track data concentrates semantic association between tracing point, make it possible to more fully understand text
The intrinsic meaning of this description.
Preferably, it includes following for searching the candidate tracks of query point q (the query point q is any one query point)
Step:
(1), it is based on the corresponding theme probability distribution of query point q, the leaf node of quaternary tree described in recursive traversal obtains excellent
First queue, the Priority Queues are ranked up according to the ascending order of mdist (q, N), and the mdist (q, N) indicates query point q and leaf
The minimum range for multiple tracing points that child node N is indicated, mdist (q, N) is smaller, and query point q and leaf node N indicates multiple
The similarity of tracing point is higher.
Preferably, the calculation formula of the mdist (q, N) is as follows:
Mdist (q, N)=λ Ds(q,N)+(1-λ)·DT(q, N),
Wherein λ ∈ [0,1] is weight parameter, DTThe specific formula for calculation of (q, N) are as follows:
Wherein DS(q, N) is the minimum boundary matrix N .rect based on leaf node N, from query point q to leaf node N
The smallest space length；DT(q, N) is the minimum theme distance of the multiple tracing points indicated from q to the leaf node N.
||TDq, N.o | | it is the theme in space and inquires
Euclidean distance between the theme distribution of point q to anchor point N.o.
(2), each leaf node in the Priority Queues is successively traversed, multiprobe LSH index (Multi- is utilized
Probe LSH Indexing) the corresponding LSH index structure of each leaf node of technology traversal, obtain the candidate of the query point
Tracing point looks into the track of the candidate tracks point comprising the query point as the candidate tracks of the query point, and by described
The a part of the candidate tracks of inquiry point as the candidate tracks of the query set Q.
For example, there are two leaf node, the first node and the second node, the sequence in Priority Queues are as follows: the second node,
First node.The tracing point under second node is traversed first with multiprobe LSH index technology, then is searched under the first node
Tracing point.
Specifically, the multiprobe LSH index technology utilizes detection sequence (carefully derived probing
Sequence), obtain and the approximate multiple Hash buckets of query point.According to the property of LSH, if we know and query point q phase
Close data are no and query point q is mapped in the same bucket, it is probably mapped in the bucket of surrounding (i.e. two buckets
Cryptographic Hash there was only little difference), target in this way be to position these buckets closed on, search neighbour's data to increase
Chance.
Each query point is inquired according to above-mentioned steps (1), (2), obtains each query point in the query set Q
Candidate tracks.
By above-mentioned implementation, it is based on the corresponding theme probability distribution of each query point, calculates each query point and semantic rail
The similarity of tracing point in mark data set, query point is characterized based on the similarity of theme distribution and semantic track data concentrates rail
Semantic association between mark point makes it possible to more fully understand the intrinsic meaning of text description.For example query specification " drinks coffee in this way
Coffee " and tracing point description " Startbuck " will be considered related because of its similar theme distribution.It can make to inquire so more acurrate.Example
As query specification " drinking coffee " and tracing point description " Startbuck " will be considered related because of its similar theme distribution.Such energy
Make to inquire more acurrate.
S13, the candidate tracks collection based on the query set Q, calculate the candidate rail of the query set Q Yu the query set Q
Mark concentrates the distance of each track, according to apart from size, concentrates each candidate tracks to carry out the candidate tracks of the query set Q
Sequence.
Specifically, semanteme track data collection a τ, a limited theme collection Z are given and includes a series of looking into for query points
Ask collection Q, shaping variable k that a user specifies (k < | τ |), the then similar inquiry in semantic track that a user oriented is intended to
(User-oriented Trajectory Similarity Query, UTSQ), returns to independent k track from τ, and this k
Trajectory distance query set Q has the smallest distance D of top-kQ(Tr)。
Preferably, the candidate tracks for calculating the query set Q and query set Q concentrate any one candidate tracks
The distance of Tr includes:
(1) each query point is calculated at a distance from each tracing point in the candidate tracks Tr.
Specifically, give a query point q containing text entry q.W and geographical location q.l, from a tracing point p to
Its distance can according between them spatial proximity and topic relativity measure, specific formula for calculation is as follows:
D (q, p)=λ DS(q,p)+(1-λ)·DT(q, p),
Wherein, λ ∈ [0,1] is that user specifies parameter to be used to adjust the weight of spatial proximity and topic similarity；DS
(q, p) is space Euclidean distance, herein equally using the sigmoid function specification distance between section [0,1]；DT(q, p),
It is DTThe simplification of (q.W, p.W) represents the theme distance between q and p text entry.
(2) according to each query point at a distance from each tracing point in the candidate tracks Tr, calculate each query point with
The distance of the candidate tracks.
Specifically, an a query point q and track Tr is given, if one of tracing point p ∈ Tr, for the track
In for any one other tracing point p', we have d (q, p)≤d (q, p'), then tracing point p can be expressed as in the track
With the maximally related tracing point of query point q (Most Relevant Point, MRP), it is defined as Tr.MRP (q).Then from most reference point
The distance between Tr.MRP (q) to tracing point q is indicated as the query point to the distance of track, under specifically can define:
Dmrp(q, Tr)=minp∈TrD (q, p),
(3) query set and the candidate tracks are calculated at a distance from the candidate tracks Tr according to each query point
The distance of Tr.
Specifically, the query set Q={ q comprising m query point is given1,q2,…qmAnd a track Tr, we define and look into
Ask the distance D of collection Q to track TrQIt (Tr) is each query point qi(i ∈ [1, m]) arrives the sum of the distance of track Tr, calculates as follows:
It can be seen that the most related point set MRPs of each query point is formed the most related point set of the inquiry in inquiry,
Tr.MRPs (Q), therefore the most related point set MRPs for finding a query set Q can be broken down into and search each inquiry in the inquiry
The most reference point MRP of point.
S14 exports result to user according to the candidate tracks collection after sequence.
Preferably, show on a user interface sequence after candidate tracks collection, the candidate tracks collection after the sequence be by
Range is from being from small to large ranked up.It thus can be by maximally related as the result is shown up front so that user is intuitive to see
Maximally related query result.
The present invention is converted to each look by the similarity measurements flow function based on theme distribution, by the description of each query point
Ask the corresponding theme probability distribution with position and time tag of point, the corresponding theme probability based on each query point point
Each query point is scanned for matching, searches the candidate rail of the query set Q by cloth with the semantic track data collection in database
Mark collection, based on the candidate tracks collection of the query set Q, the candidate tracks for calculating the query set Q and query set Q are concentrated often
The distance of a track is ranked up the candidate tracks collection of the query set Q, according to apart from size according to the candidate after sequence
Track collection exports result to user.The present invention indicates that model indicates the tracing point in query point and database using semantic track,
Tracing point and the description of the text of query point are converted into the probability distribution that is the theme, i.e., a series of master with position and time tag
Inscribe probability distribution, make it possible to more fully understand text description intrinsic meaning, and by the similarity based on theme distribution come
Their semantic association is characterized, to improve retrieval precision.
As shown in Fig. 2, the functional block diagram of the first preferred embodiment of track inquiry unit of the present invention.The track inquiry
Device 2 includes, but are not limited to one or more following module: obtaining module 20, conversion module 21, searching module 22, sequence
Module 23 and output module 24.The so-called unit of the present invention refers to that one kind can be performed by the processor of track inquiry unit 2
And the series of computation machine program segment of fixed function can be completed, storage is in memory.Function about each unit will
It is described in detail in subsequent embodiment.
The query set Q for obtaining module 20 and obtaining the description including each query point.
In alternative embodiment of the invention, the query set Q includes at least one query point.For example, user is in user
The description with multiple positions inputted on interface, needs to inquire the data of the profile matching of the multiple position, one of them
Position is described as a query point, such as the financial product of the B company under safety scientific & technical corporation, then can in the query set Q
To include two query points, query point one: description, query point two to safety scientific & technical corporation: the description to B company.
It is corresponding with position and time that the description of each query point is converted to each query point by the conversion module 21
The theme probability distribution of label.
Preferably, each query point is made of text description.The text description includes position and time tag.It utilizes
Semantic track indicates that model indicates the corresponding theme probability distribution with position and time tag of each query point.
Specifically, the semantic track indicates that model is as follows: each query point is one and is made of n spatial key
Text W, Z={ z1,z2,…,zmBe the theme collection, and the theme collection is for indicating theme belonging to n spatial key, then W pairs
The probability topic distribution TD answeredWThe middle each theme TD of correspondenceW[zi] calculation formula it is as follows:
Wherein Nw∈(WIWz)Indicating the number of keyword in W, α indicates symmetrical border, is usually arranged as 0.1, | W | it indicates in W
The number of keyword, | Z | indicate theme number in total.
Corresponding theme probability distribution of the searching module 22 based on each query point, by each query point and database
In semantic track data collection scan for matching, search the candidate tracks collection of the query set Q.
Preferably, it includes position coordinates and theme distribution information that the semantic track data, which concentrates each tracing point, is based on
Each tracing point includes that position coordinates and theme distribution information establish the hierarchical index structure including space layer and subject layer.Wherein
Space layer establishes index structure using quaternary tree, to achieve the purpose that in space layer fast convergence, and in four forks of space layer
Each leaf node of tree indicates multiple tracing points, for multiple tracing points that each leaf node indicates, establishes in subject layer
The corresponding LSH index structure of each leaf node based on position sensing Hash, in this way by using hash function by each leaf
Similar tracing point is mapped to same Hash bucket in multiple tracing points that child node indicates.
Specifically, quaternary tree index is the tree construction that location information recurrence is divided into different levels.By known range
Space is divided into four equal subspaces, and such recurrence is gone down, until the level of tree reaches certain depth or meets certain
Stop segmentation after it is required that.Quaternary tree it is relatively simple for structure, and when spatial data object distribution it is relatively uniform when, have compare
High spatial data insertion and search efficiency.Tracing point is stored in leaf node, and intermediate node and root node do not store
Tracing point.
The corresponding theme probability distribution of each query point is preferably based on to scan for matching with semantic track data collection,
The candidate tracks point of each query point is obtained, using the track of the candidate tracks point comprising each query point as each query point
Candidate tracks, using the candidate tracks of each query point as the candidate tracks of the query set Q.It is based on each query point pair in this way
The theme probability distribution answered, calculates each query point and semantic track data concentrates the similarity of tracing point, is based on theme distribution
Similarity characterize query point and semantic track data concentrates semantic association between tracing point, make it possible to more fully understand text
The intrinsic meaning of this description.
Preferably, the searching module 22 searches the time of a query point q (the inquiry q point is any one query point)
Select track the following steps are included:
(1), it is based on the corresponding theme probability distribution of query point q, the leaf node of quaternary tree described in recursive traversal obtains excellent
First queue, the Priority Queues are ranked up according to the ascending order of mdist (q, N), and the mdist (q, N) indicates query point q and leaf
The minimum range for multiple tracing points that child node N is indicated, mdist (q, N) is smaller, and query point q and leaf node N indicates multiple
The similarity of tracing point is higher.
Preferably, the calculation formula of the mdist (q, N) is as follows:
Mdist (q, N)=λ Ds(q,N)+(1-λ)·DT(q, N),
Wherein DS(q, N) is the minimum boundary matrix N .rect based on leaf node N, from query point q to leaf node N
The smallest space length；DT(q, N) is the minimum theme distance of the multiple tracing points indicated from q to the leaf node N.
Wherein λ ∈ [0,1] is weight parameter, DTThe specific formula for calculation of (q, N) are as follows:
||TDq, N.o | | it is the theme in space and inquires
Euclidean distance between the theme distribution of point q to anchor point N.o.
(2), each leaf node in the Priority Queues is successively traversed, multiprobe LSH index (Multi- is utilized
Probe LSH Indexing) the corresponding LSH index structure of each leaf node of technology traversal, obtain the candidate of the query point
Tracing point looks into the track of the candidate tracks point comprising the query point as the candidate tracks of the query point, and by described
The a part of the candidate tracks of inquiry point as the candidate tracks of the query set Q.
For example, there are two leaf node, the first node and the second node, the sequence in Priority Queues are as follows: the second node,
First node.The tracing point under second node is traversed first with multiprobe LSH index technology, then is searched under the first node
Tracing point.
Specifically, the multiprobe LSH index technology utilizes detection sequence (carefully derived probing
Sequence), obtain and the approximate multiple Hash buckets of query point.According to the property of LSH, if we know and query point q phase
Close data are no and query point q is mapped in the same bucket, it is probably mapped in the bucket of surrounding (i.e. two buckets
Cryptographic Hash there was only little difference), target in this way be to position these buckets closed on, search neighbour's data to increase
Chance.
Each query point is inquired according to above-mentioned steps (1), (2), obtains each query point in the query set Q
Candidate tracks.
By above-mentioned implementation, it is based on the corresponding theme probability distribution of each query point, calculates each query point and semantic rail
The similarity of tracing point in mark data set, query point is characterized based on the similarity of theme distribution and semantic track data concentrates rail
Semantic association between mark point makes it possible to more fully understand the intrinsic meaning of text description.For example query specification " drinks coffee in this way
Coffee " and tracing point description " Startbuck " will be considered related because of its similar theme distribution.It can make to inquire so more acurrate.
Candidate tracks collection of the sorting module 23 based on the query set Q, calculates the query set Q and the query set
The candidate tracks of Q concentrate the distance of each track, according to apart from size, concentrate each time to the candidate tracks of the query set Q
Track is selected to be ranked up.
Specifically, semanteme track data collection a τ, a limited theme collection Z are given and includes a series of looking into for query points
Ask collection Q, shaping variable k that a user specifies (k < | τ |), the then similar inquiry in semantic track that a user oriented is intended to
(User-oriented Trajectory Similarity Query, UTSQ), returns to independent k track from τ, and this k
Trajectory distance query set Q has the smallest distance D of top-kQ(Tr)。
Preferably, the candidate tracks that the sorting module 23 calculates the query set Q and query set Q are concentrated any one
The distance of a candidate tracks Tr includes:
(1) each query point is calculated at a distance from each tracing point in the candidate tracks Tr.
Specifically, give a query point q containing text entry q.W and geographical location q.l, from a tracing point p to
Its distance can according between them spatial proximity and topic relativity measure, specific formula for calculation is as follows:
D (q, p)=λ DS(q,p)+(1-λ)·DT(q, p),
Wherein, λ ∈ [0,1] is that user specifies parameter to be used to adjust the weight of spatial proximity and topic similarity；DS
(q, p) is space Euclidean distance, herein equally using the sigmoid function specification distance between section [0,1]；DT(q, p),
It is DTThe simplification of (q.W, p.W) represents the theme distance between q and p text entry.
(2) according to each query point at a distance from each tracing point in the candidate tracks Tr, calculate each query point with
The distance of the candidate tracks.
Specifically, an a query point q and track Tr is given, if one of tracing point p ∈ Tr, for the track
In for any one other tracing point p', we have d (q, p)≤d (q, p'), then tracing point p can be expressed as in the track
With the maximally related tracing point of query point q (Most Relevant Point, MRP), it is defined as Tr.MRP (q).Then from most reference point
The distance between Tr.MRP (q) to tracing point q is indicated as the query point to the distance of track, under specifically can define:
Dmrp(q, Tr)=minp∈TrD (q, p),
(3) query set and the candidate tracks are calculated at a distance from the candidate tracks Tr according to each query point
The distance of Tr.
Specifically, the query set Q={ q comprising m query point is given1,q2,…qmAnd a track Tr, we define and look into
Ask the distance D of collection Q to track TrQIt (Tr) is each query point qi(i ∈ [1, m]) arrives the sum of the distance of track Tr, calculates as follows:
It can be seen that the most related point set MRPs of each query point is formed the most related point set of the inquiry in inquiry,
Tr.MRPs (Q), therefore the most related point set MRPs for finding a query set Q can be broken down into and search each inquiry in the inquiry
The most reference point MRP of point.
Output module 24 exports result to user according to the candidate tracks collection after sequence.
Preferably, the output module 24 shows the candidate tracks collection after sequence on a user interface, after the sequence
Candidate tracks collection is ranked up from small to large according to distance.Thus can by it is maximally related as the result is shown up front for
User is intuitive to see maximally related query result.
The present invention is converted to each look by the similarity measurements flow function based on theme distribution, by the description of each query point
Ask the corresponding theme probability distribution with position and time tag of point, the corresponding theme probability based on each query point point
Each query point is scanned for matching, searches the candidate rail of the query set Q by cloth with the semantic track data collection in database
Mark collection, based on the candidate tracks collection of the query set Q, the candidate tracks for calculating the query set Q and query set Q are concentrated often
The distance of a track is ranked up the candidate tracks collection of the query set Q, according to apart from size according to the candidate after sequence
Track collection exports result to user.The present invention indicates that model indicates the tracing point in query point and database using semantic track,
Tracing point and the description of the text of query point are converted into the probability distribution that is the theme, i.e., a series of master with position and time tag
Inscribe probability distribution, make it possible to more fully understand text description intrinsic meaning, and by the similarity based on theme distribution come
Their semantic association is characterized, to improve retrieval precision.
The above-mentioned integrated unit realized in the form of software function module, can store and computer-readable deposit at one
In storage media.Above-mentioned software function module is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the present invention
The part steps of embodiment the method.
As shown in figure 3, the electronic equipment 3 includes at least one sending device 31, at least one processor 32, at least one
A processor 33, at least one reception device 34 and at least one communication bus.Wherein, the communication bus is for realizing this
Connection communication between a little components.
The electronic equipment 3 be it is a kind of can according to the instruction for being previously set or store, automatic progress numerical value calculating and/or
The equipment of information processing, hardware include but is not limited to microprocessor, specific integrated circuit (Application Specific
Integrated Circuit, ASIC), programmable gate array (Field-Programmable Gate Array, FPGA), number
Word processing device (Digital Signal Processor, DSP), embedded device etc..The electronic equipment 3 may also include network
Equipment and/or user equipment.Wherein, the network equipment includes but is not limited to single network server, multiple network servers
The server group of composition or the cloud being made of a large amount of hosts or network server for being based on cloud computing (Cloud Computing),
Wherein, cloud computing is one kind of distributed computing, a super virtual computing consisting of a loosely coupled set of computers
Machine.
The electronic equipment 3, which may be, but not limited to, any one, to pass through keyboard, touch tablet or voice-operated device with user
Etc. modes carry out the electronic product of human-computer interaction, for example, tablet computer, smart phone, personal digital assistant (Personal
Digital Assistant, PDA), intellectual wearable device, picture pick-up device, the terminals such as monitoring device.
Network locating for the electronic equipment 3 includes, but are not limited to internet, wide area network, Metropolitan Area Network (MAN), local area network, virtual
Dedicated network (Virtual Private Network, VPN) etc..
Wherein, the reception device 34 and the sending device 31 can be wired sending port, or wirelessly set
It is standby, for example including antenna assembly, for carrying out data communication with other equipment.
The memory 32 is for storing program code.The memory 32, which can be, does not have physical form in integrated circuit
The circuit with store function, such as RAM (Random-Access Memory, random access memory), FIFO (First In
First Out) etc..Alternatively, the memory 32 is also possible to the memory with physical form, such as memory bar, TF card
(Trans-flash Card), smart media card (smart media card), safe digital card (secure digital
Card), storage facilities such as flash memory cards (flash card) etc..
The processor 33 may include one or more microprocessor, digital processing unit.The processor 33 is adjustable
With the program code stored in memory 32 to execute relevant function.For example, modules described in Fig. 2 are stored in institute
The program code in memory 32 is stated, and as performed by the processor 33, to realize a kind of track querying method.The processing
Device 33 is also known as central processing unit (CPU, Central Processing Unit), is one piece of ultra-large integrated circuit, is fortune
Calculate core (Core) and control core (Control Unit).
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer instruction, the finger
It enables when the electronic equipment for being included one or more processors executes, executes electronic equipment as described in embodiment of the method above
Track querying method.
In conjunction with shown in Fig. 1 and Fig. 2, the memory 32 in the electronic equipment 3 stores multiple instruction to realize one kind
The multiple instruction can be performed to realize in track querying method, the processor 33:
Obtain the query set of the description including each query point；The description of each query point is converted into each query point pair
The theme probability distribution with position and time tag answered；Corresponding theme probability distribution based on each query point, will be every
A query point scans for matching with the semantic track data collection in database, searches the candidate tracks collection of the query set；Base
In the candidate tracks collection of the query set, the candidate tracks for calculating the query set and the query set concentrate each candidate tracks
Distance concentrate each candidate tracks to be ranked up the candidate tracks of the query set according to apart from size；After sequence
Candidate tracks collection, export result to user.
The characteristic means of present invention mentioned above can be realized by integrated circuit, and control above-mentioned of realization
The function of track querying method described in embodiment of anticipating.That is, integrated circuit of the invention is installed in the electronic equipment, make institute
It states electronic equipment to play the following functions: obtaining the query set of the description including each query point；The description of each query point is turned
It is changed to the corresponding theme probability distribution with position and time tag of each query point；Corresponding master based on each query point
Probability distribution is inscribed, each query point is scanned for matching with the semantic track data collection in database, searches the query set
Candidate tracks collection；Candidate tracks collection based on the query set, calculates the candidate tracks of the query set Yu the query set
The distance for concentrating each candidate tracks, according to apart from size, to the candidate tracks of the query set concentrate each candidate tracks into
Row sequence；According to the candidate tracks collection after sequence, result is exported to user.
Function achieved by the track querying method described in any embodiment can be transferred through integrated circuit of the invention
It is installed in the electronic equipment, plays the electronic equipment achieved by track querying method described in any embodiment
Function, this will not be detailed here.
It should be noted that for the various method embodiments described above, for simple description, therefore, it is stated as a series of
Combination of actions, but those skilled in the art should understand that, the present invention is not limited by the sequence of acts described because
According to the present invention, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art should also know
It knows, the embodiments described in the specification are all preferred embodiments, and related actions and modules is not necessarily of the invention
It is necessary.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, there is no the portion being described in detail in some embodiment
Point, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed device, it can be by another way
It realizes.For example, the apparatus embodiments described above are merely exemplary, such as the division of the unit, it is only a kind of
Logical function partition, there may be another division manner in actual implementation, such as multiple units or components can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be through some interfaces, the indirect coupling or communication connection of device or unit,
It can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in various embodiments of the present invention can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code
Medium.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although referring to before
Stating embodiment, invention is explained in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features；And these
It modifies or replaces, the range for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (9)
1. a kind of track querying method, which is characterized in that the described method includes:
Obtain the query set of the description including each query point；
The description of each query point is converted into the corresponding theme probability distribution with position and time tag of each query point；
Corresponding theme probability distribution based on each query point, by the semantic track data collection in each query point and database
It scans for matching, searches the candidate tracks of each query point, the candidate tracks of each query point are determined as the query set
Candidate tracks collection, wherein the candidate tracks of one query point q of inquiry include: based on query point q corresponding theme probability point
Cloth, the leaf node of recursive traversal quaternary tree obtain Priority Queues, and the Priority Queues is carried out according to the ascending order of mdist (q, N)
Sequence, the mdist (q, N) indicate that query point q and leaf node N indicate the minimum range of multiple tracing points, successively traverse institute
Each leaf node in Priority Queues is stated, traverses the corresponding LSH index of each leaf node using multiprobe LSH index technology
Structure obtains the candidate tracks point of the query point, and the track of the candidate tracks point comprising the query point is looked into as described in
The candidate tracks of point are ask, and using the candidate tracks of the query point as a part of the candidate tracks of the query set；
Candidate tracks collection based on the query set, the candidate tracks for calculating the query set and the query set concentrate each time
The distance for selecting track concentrates each candidate tracks to be ranked up the candidate tracks of the query set according to apart from size；
According to the candidate tracks collection after sequence, result is exported to user.
2. track querying method as described in claim 1, which is characterized in that the semanteme track data concentrates each tracing point
It include that position coordinates and theme distribution information are established including sky based on each tracing point comprising position coordinates and theme distribution information
The hierarchical index structure of interbed and subject layer, wherein space layer establishes index structure using quaternary tree, for each leaf node
The multiple tracing points indicated establish the corresponding LSH index structure of each leaf node based on position sensing Hash in subject layer.
3. track querying method as described in claim 1, which is characterized in that the calculation formula of the mdist (q, N) is as follows:
Mdist (q, N)=λ Ds(q,N)+(1-λ)·DT(q,N)；
Wherein DS(q, N) is the minimum boundary matrix N .rect based on leaf node N, most from query point q to leaf node N
Small space length, DT(q, N) from q to the leaf node N the multiple tracing points indicated minimum theme distance, λ ∈ [0,1]
For weight parameter.
4. track querying method as described in claim 1, which is characterized in that calculating the query set Q and query set Q
Candidate tracks concentrate candidate tracks apart from when, comprising:
Each query point is calculated at a distance from each tracing point in the candidate tracks Tr；
According to each query point at a distance from each tracing point in the candidate tracks Tr, each query point and the candidate are calculated
The distance of track Tr；
According to each query point at a distance from the candidate tracks Tr, calculate the query set and the candidate tracks Tr away from
From.
5. track querying method as claimed in claim 4, which is characterized in that described to calculate each query point and the candidate rail
The distance of each tracing point includes: in mark Tr
Give a query point q containing text entry q.W and geographical location q.l, from a tracing point p to query point q away from
From can according between tracing point p to query point q spatial proximity and topic relativity measure, calculation formula is as follows:
D (q, p)=λ DS(q,p)+(1-λ)·DT(q, p), wherein λ ∈ [0,1] is that user specifies parameter to be used to adjust sky
Between the degree of approach and topic similarity weight；DS(q, p) is space Euclidean distance；DT(q, p) represents the master between q and p text entry
Inscribe distance.
6. track querying method as claimed in claim 4, which is characterized in that described according in each query point and candidate tracks
The distance of each tracing point, calculate each query point includes: at a distance from the candidate tracks Tr
Given an a query point q and track Tr, one of tracing point p ∈ Tr, for any one of other in the track
For tracing point p', there is d (q, p)≤d (q, p'), then tracing point p is expressed as in the track and the maximally related tracing point of query point q
Tr.MRP (q) is then indicated as the query point to rail the distance between from maximally related tracing point Tr.MRP (q) to tracing point q
The distance of mark.
7. track querying method as claimed in claim 4, which is characterized in that described according to each query point and the candidate rail
The distance of mark, calculate the query set includes: at a distance from the candidate tracks Tr
The given inquiry Q={ q comprising m inquiry point set1,q2,…qmAnd a track Tr, inquire the distance D of Q to track TrQ
It (Tr) is each query point qi(i ∈ [1, m]) arrives the sum of the distance of track Tr, calculates as follows:
The most related point set MRPs of each query point is formed the most related point set Tr.MRPs (Q) of the inquiry in inquiry.
8. a kind of electronic equipment, which is characterized in that the electronic equipment includes memory and processor, and the memory is for depositing
At least one instruction is stored up, the processor is for executing at least one described instruction to realize such as any one of claims 1 to 7
The track querying method.
9. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has at least one
Instruction, at least one described instruction realize the track issuer as described in any one of claims 1 to 7 when being executed by processor
Method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810303902.0A CN108536813B (en) | 2018-04-04 | 2018-04-04 | Track querying method, electronic equipment and storage medium |
| PCT/CN2018/100149 WO2019192120A1 (en) | 2018-04-04 | 2018-08-13 | Track query method, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810303902.0A CN108536813B (en) | 2018-04-04 | 2018-04-04 | Track querying method, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108536813A CN108536813A (en) | 2018-09-14 |
| CN108536813B true CN108536813B (en) | 2019-07-16 |
Family
ID=63482647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810303902.0A Active CN108536813B (en) | 2018-04-04 | 2018-04-04 | Track querying method, electronic equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108536813B (en) |
| WO (1) | WO2019192120A1 (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109492150B (en) * | 2018-10-30 | 2021-07-27 | 石家庄铁道大学 | Reverse nearest neighbor query method and device based on semantic track big data |
| CN111241217B (en) * | 2018-11-29 | 2023-05-30 | 阿里巴巴集团控股有限公司 | Data processing method, device and system |
| CN110543457A (en) * | 2019-09-11 | 2019-12-06 | 北京明略软件系统有限公司 | Track type document processing method and device, storage medium and electronic device |
| CN111221353B (en) * | 2020-04-16 | 2020-08-14 | 上海特金信息科技有限公司 | Unmanned aerial vehicle flight trajectory processing method and device, electronic equipment and storage medium |
| CN112487256B (en) * | 2020-12-10 | 2024-05-24 | 中国移动通信集团江苏有限公司 | Object query method, device, equipment and storage medium |
| CN112597190A (en) * | 2020-12-28 | 2021-04-02 | 京东城市(北京)数字科技有限公司 | Point neighbor track query method and device, electronic equipment and readable storage medium |
| CN114117260B (en) * | 2021-12-02 | 2022-09-23 | 中国人民解放军国防科技大学 | Spatiotemporal trajectory indexing and query processing method, device, equipment and medium |
| CN115510303B (en) * | 2022-11-02 | 2023-04-14 | 深圳市瑞达飞行科技有限公司 | Method and system for identifying off-field program, computer equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102214215A (en) * | 2011-06-07 | 2011-10-12 | 陆嘉恒 | Rapid reverse nearest neighbour search method based on text information |
| CN104834679A (en) * | 2015-04-14 | 2015-08-12 | 苏州大学 | Representation and inquiry method of behavior track and device therefor |
-
2018
- 2018-04-04 CN CN201810303902.0A patent/CN108536813B/en active Active
- 2018-08-13 WO PCT/CN2018/100149 patent/WO2019192120A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102214215A (en) * | 2011-06-07 | 2011-10-12 | 陆嘉恒 | Rapid reverse nearest neighbour search method based on text information |
| CN104834679A (en) * | 2015-04-14 | 2015-08-12 | 苏州大学 | Representation and inquiry method of behavior track and device therefor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108536813A (en) | 2018-09-14 |
| WO2019192120A1 (en) | 2019-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108536813B (en) | Track querying method, electronic equipment and storage medium | |
| CN107688823B (en) | A kind of characteristics of image acquisition methods and device, electronic equipment | |
| CN104123332B (en) | The display methods and device of search result | |
| CN102915347B (en) | A kind of distributed traffic clustering method and system | |
| CN110234085B (en) | Indoor position fingerprint map generation method and system based on anti-migration network | |
| CN106128100B (en) | A kind of short-term traffic flow forecast method based on Spark platform | |
| JP6070956B2 (en) | Similarity detection apparatus and directivity vicinity detection method | |
| CN107346326A (en) | For generating the method and system of neural network model | |
| CN114549874B (en) | Training method of multi-target image-text matching model, image-text retrieval method and device | |
| CN109993748A (en) | A kind of three-dimensional grid method for segmenting objects based on points cloud processing network | |
| CN110019876A (en) | Data query method, electronic equipment and storage medium | |
| CN110019650A (en) | Method, apparatus, storage medium and the electronic equipment of search associational word are provided | |
| JP7393475B2 (en) | Methods, apparatus, systems, electronic devices, computer readable storage media and computer programs for retrieving images | |
| CN110191183A (en) | Accurate intelligent method for pushing, system, device and computer readable storage medium | |
| CN107315984B (en) | Pedestrian retrieval method and device | |
| Xuan et al. | Constrained range search query processing on road networks | |
| CN108053424A (en) | Method for tracking target, device, electronic equipment and storage medium | |
| CN108509545A (en) | A kind of comment processing method and system of article | |
| Chen et al. | A deep learning based fast-flux and CDN domain names recognition method | |
| Kaur et al. | SIMHAR-smart distributed web crawler for the hidden web using SIM+ hash and redis server | |
| Sun et al. | On efficient aggregate nearest neighbor query processing in road networks | |
| CN110019400A (en) | Date storage method, electronic equipment and storage medium | |
| CN109284409A (en) | Picture group geographic positioning based on extensive streetscape data | |
| CN107368923A (en) | Sight spot temperature Forecasting Methodology and device | |
| Bhattacharya et al. | LAWRA: a layered wrapper feature selection approach for network attack detection |
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 |