CN106951455A - A kind of similar track analysis system and its analysis method - Google Patents
A kind of similar track analysis system and its analysis method Download PDFInfo
- Publication number
- CN106951455A CN106951455A CN201710101803.XA CN201710101803A CN106951455A CN 106951455 A CN106951455 A CN 106951455A CN 201710101803 A CN201710101803 A CN 201710101803A CN 106951455 A CN106951455 A CN 106951455A
- Authority
- CN
- China
- Prior art keywords
- track
- analysis
- similarity
- data
- similar
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
-
- 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)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of similar track analysis system, including several mobile terminals and storage analysis processor, wherein storage analysis processor includes user interface layer, Business Logic and data storage layer；A kind of analysis method of similar track analysis system as described above, including the definition of 1, parameter, adjust most bad value and generate dynamic similarity analysis model；2nd, the trail file that user submits is obtained；3rd, track data is pre-processed；4th, similarity between track is calculated；5th, tracing point similar between acquisition track, is compared and is analyzed again the information similitude, obtains similarity highest most picture point；6th, similar orbit segment is searched out；7th, encapsulate analysis result and be delivered to UI interfaces；8th, result is showed by user in visual mode, facilitates user to check analysis result.The present invention can clear data between noise, calculating speed is fast, efficiency high.
Description
Technical field
The invention belongs to data analysis field, more particularly to a kind of similar track analysis system and its analysis method.
Background technology
With the fast development of the position such as GPS device, sensor network, satellite and radio communication acquisition technique, all kinds of shiftings
Dynamic object generates large-scale track data.Track data generally includes track sets and tracing point, wherein：Tracing point is note
Record the atomic data of track, one group of record that it is made up of longitude, latitude；Track sets are by some rails comprising longitude, latitude
Mark point is constituted.By the trajectory analysis to mobile object, many highly useful information can be known：(1) these space-times are utilized
Data can be every Track Pick-up its respective mobile attribute, and this is for studying in the Information Mobile Service based on geographical location information
The itinerary for excavating anonymous is helpful；Us can also be helped to excavate interest of the user within the past period to live
Dynamic region.(2) track data can be additionally used in the optimization and design of traffic route.The important work that taxi is migrated as Urban population
Tool, taxi driver is often familiar to road conditions, and their traffic route can be considered as the optimal road between 2 points
Footpath (classical path), therefore, can find most fast traffic route by hiring out the statistical law of wheel paths.(3) to multi-trace
Processing is calculated, and can search for out most like orbit segment and tracing point, and civic communal facility is repaiied for its planning to city
Have reference role.Therefore, how the trace information of effective management and use mobile object turns into the focus studied at present
One of.
It is not difficult to find out from some above-mentioned demands, the similarity of research track data, realizes that similar track analysis system is
Significantly, there can be technical support to share-car application in actual life, for the detection of maximum flow place and track
It is all very helpful in terms of analysis.Received in the GeoLife projects that Geolife data sets are provided from Microsoft Research, Asia
The track data collection collected, the data set is adopted comprising GPS track of 165 users during in April, 2007 in August, 2009
Sample, is one of significant data collection of industry research track data analysis.Therefore, the present invention have studied based on Geolife data sets
A kind of construction method of similar track analysis system.
The content of the invention
Goal of the invention：For problems of the prior art, the present invention, which is provided, a kind of can return to the similar of two tracks
Spend result of calculation；Most like orbit segment between track can be found out, can clearly be represented on map；Most like rail can be searched out
Mark section, returns to the similar result of calculation of most like orbit segment, and can on map the similar track analysis system of distinct mark and
Its analysis method.
Technical scheme：In order to solve the above technical problems, the present invention provides a kind of side measured similitude track
Method；And to the preprocess method of track data, novelty carries out outlier detection using clustering method.
A kind of similar track analysis system, including several mobile terminals and storage analysis processor, wherein storage point
Analysing processor includes user interface layer, Business Logic and data storage layer；
Data storage layer, the four type data needed for storage system：Track abstract structure, similarity analysis data,
Visualization resource and data-interface and agreement；
Track abstract structure is the data structure of track data in data Geolife data sets, including track definition and rail
The definition of mark point；Similarity analysis data are the definition of scale model abstract data structure, and the needs in progress similarity analysis
Some constant definitions, such as most bad value；When visualization resource is that track is shown in Baidu map, the number being packaged to track
According to type, the resource such as the most like orbit segment of mark and the data structure of tracing point is needed on map；Data-interface and agreement are
Customized special access mode when Business Logic is interacted with data storage layer, and the various symbols in access process
Parsing.
Business Logic, for carrying out data acquisition, data prediction, similarity analysis and abnormality processing；
Data acquisition includes the reading to multiple csv files in track in single csv file and file, is respectively used to double
Track and the similarity analysis of multi-trace；Data prediction includes missing values processing, outlier detection, track cutting and trajectory synchronization
Adjustment.The method that missing values processing is used is linear interpolation method and mean value method, and outlier detection is to utilize K-means cluster sides
Method is realized, participates in using tracing point as elementary cell calculating, and track cutting is easy to improve phase using the method cut by flex point
Like the granularity of analysis, it is accurate to the calculating of orbit segment and compares；Trajectory synchronization adjustment is to carry out unified adjustment to track points；Phase
Like the analysis phase, scale model is initialized, the similarity between track is then calculated, is then searched and is most like tracing point and looks into
Look for and be most like orbit segment.Abnormality processing is that the exception captured is classified during any calculating, and according to exception
Classification selects different processing methods.
User interface layer, for providing the interactive interface with user, the result of calculation of service logic layer function is presented to
User.
User can carry out different operations by the man-machine interaction with computer, and such as file is submitted, track visualization, tune
Save most bad value, different tracks similarity etc..
A kind of analysis method of similar track analysis system as described above, comprises the following steps：
Step one：Parameter is defined, and the weights of parameter are changed according to user's request, adjusts most bad value, generation dynamic similarity point
Analyse model
Step 2：The trail file for obtaining user's submission is csv file, parses track data and stores data abstraction,
Track is numbered, tracing point ID is added；
Step 3：Track data pretreatment includes missing values processing, track cutting and outlier detection；
Step 4：Track data after input processing, utilizes similarity between similarity analysis model calculating track；
Step 5：Similar tracing point between acquisition track, is compared and is analyzed again the information similitude, is obtained
Similarity highest most picture point；
Step 6：Similar orbit segment is searched out, the similitude of secondary analysis orbit segment returns to similarity highest and is most like
Orbit segment；；
Step 7：Encapsulation analysis result is simultaneously delivered to UI interfaces；
Step 8：Result is showed by user in visual mode, facilitates user to check analysis result.
Further, in the step 4 using similarity analysis model calculate track between similarity comprise the following steps that：
First, similarity analysis is carried out to track based on track characteristic value, track characteristic value is closed on including different distance, tracing point between track
Degree, shape difference degree；Then using dynamic time warping (DTW) two tracks are carried out with the matching of corresponding points；Again to matching
Tracing point is recalled, and carries out statistical analysis, analyzes characteristic value：Sequence difference between actual range, match point between match point；Its
Actual range analysis is the embodiment of different distance and shape difference degree between track between middle match point, and the distance is based on longitude and latitude
Actual distance of curved surface；The distance set calculated is calculated variance to measure its trajectory shape similarity；3 amounts more than
Analysis the similarity final track quantified and normalized.This model calculates trajectory distance, backtracking using DTW
The tracing point matched somebody with somebody, actual distance of curved surface based on longitude and latitude calculates the actual range between match point, using statistical analysis and counts
The coefficient of variation is calculated, track is calculated sequentially using EditDistance, to 3 similarity factors (trajectory distance, trajectory shape, rails
Mark is sequentially) quantified and normalized, ask for track similarity.
Further, comprising the following steps that for similarity highest most picture point is obtained in the step 5：Calculate tracing point
With the Euclidean distance between point, m*n matrixes are formed, track point sequence in small distance is found<p1,p2>, similar point set P is formed,
The track order difference of each pair point and the position in track are analyzed again, most like tracing point is found out, and are most picture point.
Further, generation is most like comprising the following steps that for orbit segment in the step 6：According to DTW distances, dynamic is raw
Into the similar thresholding of tracing point, LCSS is carried out the search of similar track section based on the threshold value, searches out the similar orbit segment of multistage,
Carry out accurate Similarity Measure again, similarity highest is to be most like orbit segment between two tracks.
Compared with prior art, the advantage of the invention is that：User can carry out mould according to oneself demand and type of gesture
Type is adjusted and defined, and in place, the noise between clearing data, calculating speed is fast, and efficiency high, effect of visualization is good for data prediction,
Represent last result of calculation to user in the way of picture.
The present invention can return to the Similarity Measure result of two tracks；Most like orbit segment between track, energy can be found out
Clearly represent on map；It can search out most like orbit segment, return to the similar result of calculation of most like orbit segment, and can be
Distinct mark on map.
Brief description of the drawings
Fig. 1 is structural representation of the invention；
Fig. 2 is business process map of the invention；
Fig. 3 is overview flow chart of the invention.
Embodiment
With reference to the accompanying drawings and detailed description, the present invention is furture elucidated.
A kind of similar track analysis system as shown in Figure 1, including several mobile terminals and storage analysis processor, its
Middle storage analysis processor includes user interface layer, Business Logic and data storage layer；
Data storage layer, the four type data needed for storage system：Track abstract structure, similarity analysis data,
Visualization resource and data-interface and agreement；
Track abstract structure is the data structure of track data in data Geolife data sets, including track definition and rail
The definition of mark point；Similarity analysis data are the definition of scale model abstract data structure, and the needs in progress similarity analysis
Some constant definitions, such as most bad value；When visualization resource is that track is shown in Baidu map, the number being packaged to track
According to type, the resource such as the most like orbit segment of mark and the data structure of tracing point is needed on map；Data-interface and agreement are
Customized special access mode when Business Logic is interacted with data storage layer, and the various symbols in access process
Parsing.
Business Logic, for carrying out data acquisition, data prediction, similarity analysis and abnormality processing；
Data acquisition includes the reading to multiple csv files in track in single csv file and file, is respectively used to double
Track and the similarity analysis of multi-trace；Data prediction includes missing values processing, outlier detection, track cutting and trajectory synchronization
Adjustment.The method that missing values processing is used is linear interpolation method and mean value method, and outlier detection is to utilize K-means cluster sides
Method is realized, participates in using tracing point as elementary cell calculating, and track cutting is easy to improve phase using the method cut by flex point
Like the granularity of analysis, it is accurate to the calculating of orbit segment and compares；Trajectory synchronization adjustment is to carry out unified adjustment to track points；Phase
Like the analysis phase, scale model is initialized, the similarity between track is then calculated, is then searched and is most like tracing point and looks into
Look for and be most like orbit segment.Abnormality processing is that the exception captured is classified during any calculating, and according to exception
Classification selects different processing methods.
User interface layer, for providing the interactive interface with user, the result of calculation of service logic layer function is presented to
User.
User can carry out different operations by the man-machine interaction with computer, and such as file is submitted, track visualization, tune
Save most bad value, different tracks similarity etc..
As shown in Fig. 2 describing the concrete function of 3 modules of the Business Logic of application in detail, from top to bottom mainly
It is divided into three layers：Data acquisition, data prediction and similarity analysis.Data acquisition is to be submitted for track data file and similar
The weights of degree factor are set；Data prediction is that the processing such as missing values and scale model adjustment are carried out to track data；Similar point
Analysis is then calculating similarity, most picture point and is most like orbit segment and the Visual Implementation.
As shown in figure 3, clicking to enter after application, according to user's request, similarity factor weights are adjusted, each is can control
Ratio of the similar factors in similarity analysis；Trail file is submitted, system can obtain the track data in file, presented a paper point
For target trajectory file and test trails file, test file is the file for needing to carry out similarity analysis with target trajectory file；
The data got are carried out missing values processing, outlier detection, track cutting and trajectory synchronization by data prediction；It will locate
The data input managed is to similarity analysis module, and (system is parsed the similarity for calculating between track to the trail file of submission
And storage, according to submitting parameter dynamically to adjust similarity analysis model, calculate the similarity between track, according to three similarities because
Son carrys out COMPREHENSIVE CALCULATING similarity, and using distance between DTW calculating track, EditDistance calculates track sequentially, and variation lines
Count to reflect trajectory shape, quantified according to weights and most bad value and normalized)；Analyze the track order difference of tracing point
With the position in track, most like tracing point between acquisition track；LCSS is carried out similar in the threshold value based on DTW distances
Orbit segment is searched for, then carries out accurate Similarity Measure, and it is to be most like orbit segment to find out similarity highest orbit segment；Encapsulation knot
Shu Hou, returns to user and shows the page；Click on track visualization, can on map show map, check track trend and track
Shape, to most picture point and orbit segment can be most like be marked on map.
Embodiments of the invention is the foregoing is only, is not intended to limit the invention.All principles in the present invention
Within, the equivalent made should be included within the scope of the present invention.The content category that the present invention is not elaborated
In prior art known to this professional domain technical staff.
Claims (5)
1. a kind of similar track analysis system, it is characterised in that including several mobile terminals and storage analysis processor, its
Middle storage analysis processor includes user interface layer, Business Logic and data storage layer；
Data storage layer, the four type data needed for storage system：It is track abstract structure, similarity analysis data, visual
Change resource and data-interface and agreement；
Business Logic, for carrying out data acquisition, data prediction, similarity analysis and abnormality processing；
User interface layer, for providing the interactive interface with user, user is presented to by the result of calculation of service logic layer function.
User can carry out different operations by the man-machine interaction with computer, and primary operational is submitted including file, and track is visual
Change, adjust most bad value and different tracks similarity.
2. a kind of analysis method of similar track analysis system as claimed in claim 1, it is characterised in that：Including following step
Suddenly：
Step one：Parameter is defined, and the weights of parameter are changed according to user's request, adjusts most bad value, generation dynamic similarity analysis mould
Type
Step 2：The trail file for obtaining user's submission is csv file, parses track data and stores data abstraction, to rail
Mark is numbered, and adds tracing point ID；
Step 3：Track data pretreatment includes missing values processing, track cutting and outlier detection；
Step 4：Track data after input processing, utilizes similarity between similarity analysis model calculating track；
Step 5：Similar tracing point between acquisition track, is compared and is analyzed again the information similitude, is obtained similar
Spend highest most picture point；
Step 6：Similar orbit segment is searched out, the similitude of secondary analysis orbit segment returns to similarity highest and is most like track
Section；；
Step 7：Encapsulation analysis result is simultaneously delivered to UI interfaces；
Step 8：Result is showed by user in visual mode, facilitates user to check analysis result.
3. the analysis method of similar track analysis system according to claim 2, it is characterised in that：It is sharp in the step 4
Similarity is comprised the following steps that between calculating track with similarity analysis model：First, phase is carried out to track based on track characteristic value
Like analyzing, track characteristic value includes different distance, tracing point proximity, shape difference degree between track；Then dynamic time is utilized
Regular (DTW) carries out the matching of corresponding points to two tracks；The tracing point of matching is recalled again, and carries out statistical analysis, point
Analyse characteristic value：Sequence difference between actual range, match point between match point；Actual range analysis is track wherein between match point
Between different distance and shape difference degree embodiment, the distance be the actual distance of curved surface based on longitude and latitude；To the distance calculated
Set calculates variance to measure its trajectory shape similarity；The analysis of 3 amounts is come the similarity final track more than
Quantified and normalized.
4. the analysis method of similar track analysis system according to claim 2, it is characterised in that：Obtained in the step 5
Take comprising the following steps that for similarity highest most picture point：The Euclidean distance of track between points is calculated, m*n matrixes are formed,
Find track point sequence in small distance<p1,p2>, form similar point set P, then analyze each pair point track order difference and
Position in track, finds out most like tracing point, is most picture point.
5. the analysis method of similar track analysis system according to claim 2, it is characterised in that：It is raw in the step 6
Into being most like comprising the following steps that for orbit segment：According to DTW distances, the similar thresholding of dynamic generation tracing point, LCSS is based on the threshold
Value carries out the search of similar track section, searches out the similar orbit segment of multistage, then carries out accurate Similarity Measure, and similarity is most
Orbit segment is most like between high as two tracks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710101803.XA CN106951455A (en) | 2017-02-24 | 2017-02-24 | A kind of similar track analysis system and its analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710101803.XA CN106951455A (en) | 2017-02-24 | 2017-02-24 | A kind of similar track analysis system and its analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106951455A true CN106951455A (en) | 2017-07-14 |
Family
ID=59468323
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710101803.XA Pending CN106951455A (en) | 2017-02-24 | 2017-02-24 | A kind of similar track analysis system and its analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106951455A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665289A (en) * | 2017-11-17 | 2018-02-06 | 广州汇智通信技术有限公司 | The processing method and system of a kind of carrier data |
| CN108151745A (en) * | 2017-12-25 | 2018-06-12 | 千寻位置网络有限公司 | NMEA tracks difference automatically analyze and identification method |
| CN109635867A (en) * | 2018-12-10 | 2019-04-16 | 合肥工业大学 | For measuring the method and system of the mobile target trajectory similitude in ocean |
| CN109977108A (en) * | 2019-04-03 | 2019-07-05 | 深圳市甲易科技有限公司 | A kind of a variety of track collision analysis methods in Behavior-based control track library |
| CN110059141A (en) * | 2019-04-22 | 2019-07-26 | 珠海网博信息科技股份有限公司 | A method of relationship analysis is carried out to different acquisition feature by log track |
| CN110702131A (en) * | 2019-09-06 | 2020-01-17 | 杭州飞步科技有限公司 | Data processing method and device |
| CN112557072A (en) * | 2020-11-12 | 2021-03-26 | 中国煤炭科工集团太原研究院有限公司 | Method and device for calibrating spatial degree of freedom of cantilever of excavating equipment |
| CN113657177A (en) * | 2021-07-22 | 2021-11-16 | 北京百度网讯科技有限公司 | Data processing method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101923549A (en) * | 2009-07-29 | 2010-12-22 | 北京航天理想科技有限公司 | User-defined visual intelligent track clue analytical system and establishing method |
| CN102435197A (en) * | 2011-08-05 | 2012-05-02 | 刘建勋 | MBR (Master Boot Record)-based GPS (Global Position System) track map matching method |
| CN102722541A (en) * | 2012-05-23 | 2012-10-10 | 中国科学院计算技术研究所 | Method and system for calculating space-time locus similarity |
| CN103593361A (en) * | 2012-08-14 | 2014-02-19 | 中国科学院沈阳自动化研究所 | Movement space-time trajectory analysis method in sense network environment |
-
2017
- 2017-02-24 CN CN201710101803.XA patent/CN106951455A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101923549A (en) * | 2009-07-29 | 2010-12-22 | 北京航天理想科技有限公司 | User-defined visual intelligent track clue analytical system and establishing method |
| CN102435197A (en) * | 2011-08-05 | 2012-05-02 | 刘建勋 | MBR (Master Boot Record)-based GPS (Global Position System) track map matching method |
| CN102722541A (en) * | 2012-05-23 | 2012-10-10 | 中国科学院计算技术研究所 | Method and system for calculating space-time locus similarity |
| CN103593361A (en) * | 2012-08-14 | 2014-02-19 | 中国科学院沈阳自动化研究所 | Movement space-time trajectory analysis method in sense network environment |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665289A (en) * | 2017-11-17 | 2018-02-06 | 广州汇智通信技术有限公司 | The processing method and system of a kind of carrier data |
| CN107665289B (en) * | 2017-11-17 | 2020-12-08 | 广州汇智通信技术有限公司 | Operator data processing method and system |
| CN108151745A (en) * | 2017-12-25 | 2018-06-12 | 千寻位置网络有限公司 | NMEA tracks difference automatically analyze and identification method |
| CN109635867A (en) * | 2018-12-10 | 2019-04-16 | 合肥工业大学 | For measuring the method and system of the mobile target trajectory similitude in ocean |
| CN109635867B (en) * | 2018-12-10 | 2022-11-08 | 合肥工业大学 | Method and system for measuring ocean moving target track similarity |
| CN109977108A (en) * | 2019-04-03 | 2019-07-05 | 深圳市甲易科技有限公司 | A kind of a variety of track collision analysis methods in Behavior-based control track library |
| CN109977108B (en) * | 2019-04-03 | 2021-04-27 | 深圳市甲易科技有限公司 | Behavior trajectory library-based multi-trajectory collision analysis method |
| CN110059141A (en) * | 2019-04-22 | 2019-07-26 | 珠海网博信息科技股份有限公司 | A method of relationship analysis is carried out to different acquisition feature by log track |
| CN110702131A (en) * | 2019-09-06 | 2020-01-17 | 杭州飞步科技有限公司 | Data processing method and device |
| CN110702131B (en) * | 2019-09-06 | 2021-06-04 | 杭州飞步科技有限公司 | Data processing method and device |
| CN112557072A (en) * | 2020-11-12 | 2021-03-26 | 中国煤炭科工集团太原研究院有限公司 | Method and device for calibrating spatial degree of freedom of cantilever of excavating equipment |
| CN113657177A (en) * | 2021-07-22 | 2021-11-16 | 北京百度网讯科技有限公司 | Data processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951455A (en) | A kind of similar track analysis system and its analysis method | |
| Su et al. | A survey of trajectory distance measures and performance evaluation | |
| CN110245981B (en) | Crowd type identification method based on mobile phone signaling data | |
| Yao et al. | Sensing spatial distribution of urban land use by integrating points-of-interest and Google Word2Vec model | |
| Zhong et al. | Detecting the dynamics of urban structure through spatial network analysis | |
| CN110472066B (en) | Construction method of urban geographic semantic knowledge map | |
| CN105045858B (en) | Taxi pickup point based on ballot recommends method | |
| Wu et al. | Inferring demographics from human trajectories and geographical context | |
| Zhang et al. | The Traj2Vec model to quantify residents’ spatial trajectories and estimate the proportions of urban land-use types | |
| Ahmed et al. | Map construction algorithms | |
| US20150363700A1 (en) | Discovering Functional Groups of an Area | |
| Ghaemi et al. | LaSVM-based big data learning system for dynamic prediction of air pollution in Tehran | |
| WO2014194480A1 (en) | Air quality inference using multiple data sources | |
| CN103646070A (en) | Data processing method and device for search engine | |
| CN108737492A (en) | A method of the navigation based on big data system and location-based service | |
| CN116681176B (en) | Traffic flow prediction method based on clustering and heterogeneous graph neural network | |
| Ren et al. | Long-Term Preservation of Electronic Record Based on Digital Continuity in Smart Cities. | |
| Sun et al. | Identifying tourists and locals by K-means clustering method from mobile phone signaling data | |
| Bachir et al. | Combining bayesian inference and clustering for transport mode detection from sparse and noisy geolocation data | |
| John et al. | Deriving incline values for street networks from voluntarily collected GPS traces | |
| Wu et al. | Research themes of geographical information science during 1991–2020: a retrospective bibliometric analysis | |
| Cai et al. | Discovery of urban functional regions based on Node2vec | |
| Liao et al. | Fusing geographic information into latent factor model for pick-up region recommendation | |
| Dai et al. | Context-based moving object trajectory uncertainty reduction and ranking in road network | |
| Fan et al. | Towards POI-based large-scale land use modeling: spatial scale, semantic granularity, and geographic context |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170714 |
|
| RJ01 | Rejection of invention patent application after publication |