CN106599234A - Data visualization processing method and system based on multidimensional identification - Google Patents
Data visualization processing method and system based on multidimensional identification Download PDFInfo
- Publication number
- CN106599234A CN106599234A CN201611181383.2A CN201611181383A CN106599234A CN 106599234 A CN106599234 A CN 106599234A CN 201611181383 A CN201611181383 A CN 201611181383A CN 106599234 A CN106599234 A CN 106599234A
- Authority
- CN
- China
- Prior art keywords
- data
- multidimensional
- dimension
- valid
- visualization processing
- 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
- 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
- G06F16/26—Visual data mining; Browsing structured data
-
- 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
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
Abstract
The invention discloses a data visualization processing method and system based on multidimensional identification. The method comprises the steps that client big data relevant to a selected main body is collected through the Internet technology; the client big data is imported into a distributed database, the imported data is sorted out and classified, meanwhile, data screening and repetition removing are performed, and valid data is obtained after useless and invalid data is eliminated; the valid data is subjected to dimension reduction processing to obtain low-dimension data; an HTML<canvas>; label is adopted to realize image drawing for the low-dimension data by calling an API of canvas through JavaScript. The data at a ten-million order is subjected to simple visualized data graph display processing, a data analysis and processing method combining a multidimensional scaling method and data visualization is realized, the data is processed into a simple data graph through the multidimensional scaling method and visualization, and therefore existing data analysis and management and periodical data prediction in a later period can be visually promoted.
Description
Technical field
The present invention relates to the analysis of data, processing technology field, and in particular to a kind of data visualization identified based on multidimensional
Change processing method and system.
Background technology
Improving, society also to be adapted to the demand in epoch in development, data visualization to science and technology, except will in data processing and
In terms of data display outside the lower foot time, it is also to be stressed that function ease for use and user-friendly for operation, and should not have too high study threshold,
Except technical staff, allow more business personnels to will appreciate that data platform, understand data visualization.The application of data visualization
Value, its multiformity and the performance many practitioners of gravitational attraction, and each link in its production process has powerful specialty
Background is supported.Either dynamic or the visualized graphs of static state, have all built new bridge for us, and let us can see clearly generation
Boundary actually, find panoramic relation, impression is centered around the information change of ours at one's side all the time, moreover it is possible to which let us is managed
The things excavated is difficult under solution other forms.Under big data historical background, data analysiss need the support of big data just valuable
Value.Visual Chart instrument is in the big data epoch, it appears particularly critical.
Data visualization be all and data analysis function combination, data analysiss need again data access integration, data processing,
The data functions such as ETL, develop into one-stop big data analysis platform.
At present, data analysiss finally obtain a reflection sample parent mostly using cluster analyses by similar sample group
The pedigree diagram of thin relation.Cluster analyses are more simple and easy to do, but, the shortcoming of cluster analyses be by the sample of some higher-dimensions by force
Include in an one-dimensional hierarchical classification, usually simplify the relation between primary sample, or even distortion sometimes.
In addition, the method for data analysiss also has multidimensional scaling, its method is by several higher-dimension object of study, approximate
Under meaning, from higher-dimension yojan in a space compared with low-dimensional, and seek an optimal Spatial Dimension and locus (such as
2 dimensions or 3-dimensional) and still keep the primitive relation of each object of study data.
And big data is analyzed by multidimensional scaling, it has been a more ripe technology, in related field
There is a very good effect, but it is very rare that at present multidimensional scaling is combined with data visualization, and implement and also have
Many difficult points need to capture.
The content of the invention
A kind of above-mentioned technical problem in order to solve prior art presence of the invention, there is provided data identified based on multidimensional
Visible processing method and system, it is achieved thereby that multidimensional label method is combined with data visualization is analyzed place to data
Reason, facilitates the data prediction of available data analysis, data management and later stage.
For achieving the above object, the invention provides it is a kind of based on multidimensional identify data visualization processing method, including
Following steps：
The client big data related to selected main body is gathered by Internet technology；
Client big data is imported to into distributed data base, and the data to importing carry out arrangement classification, while carrying out
Data screening duplicate removal, and valid data are obtained after excluding useless invalid data；
Valid data are carried out into the process of dimensionality reduction degree and obtains low-dimensional degrees of data；
Using HTML 5<canvas>Label, calls the API of canvas to realize to low-dimensional degrees of data by JavaScript
Image rendering so that the data with ten million magnitude are processed by the displaying of simple visualization datagram.
As the preferred technical solution of the present invention, valid data carried out dimensionality reduction degree process to obtain low-dimensional degrees of data and specifically wrap
Include following steps：
Target is defined, it is determined that the target closely related with selected theme；
Using multidimensional label method, the dimension of valid data is reduced；
With minimum dimension fitting output data in the way of space diagram.
As the preferred technical solution of the present invention, using multidimensional label method, reduce valid data dimension specifically include with
Lower step：
Valid data are divided by different dimensions；
First data are obtained to valid data classification by timing node；
Second data are obtained to valid data classification by action node；
Sorted second data are carried out into row's point.
As the preferred technical solution of the present invention, using Principal Component Analysis algorithms to data
Dimension carries out reduction process.
Used as the preferred technical solution of the present invention, the computational methods of Principal Component Analysis algorithms are such as
Under：
If n-dimensional vector w is a change in coordinate axis direction of target subspace, the change in coordinate axis direction is referred to as map vector, maximum
Change the variance equation below after data mapping：
Wherein m is the number of data instance, and xi is that the vector table of data instance i reaches,It is the average of all data instances
Vector；
W is the matrix comprising all map vectors for column vector, through linear algebraic transformation, obtains following optimization aim letter
Number：
The wherein mark of tr representing matrixs, A is data covariance matrix.
Present invention also offers a kind of data visualization processing system identified based on multidimensional, including：
Data acquisition module, for gathering the client big data related to selected main body by Internet technology；
Data import modul, for client big data is imported to distributed data base, and the data to importing are carried out
Arrange and sort out, while carrying out data screening duplicate removal, and valid data are obtained after excluding useless invalid data；
Data multidimensional identifies processing module, obtains low-dimensional degrees of data for valid data are carried out drop latitude process；And
Data visualization processing module, for using HTML 5<canvas>Label, is called by JavaScript
The API of canvas realizes the image rendering to low latitudes data, so that the data with ten million magnitude are by simple visual
Change datagram displaying to process.
Used as the preferred technical solution of the present invention, the data multidimensional scale processing module is specifically included：
Target definition unit, defines target, it is determined that the target closely related with selected theme；
Dimension-reduction treatment unit, using multidimensional label method, reduces the dimension of valid data；And
Dimension determining unit, with minimum dimension fitting output data in the way of space diagram.
Used as the preferred technical solution of the present invention, dimension-reduction treatment unit is specifically included：
Dimension divides subelement, for dividing to valid data by different dimensions；
Timing node classification subelement, for obtaining first data to valid data classification by timing node；
Action node-classification subelement, for obtaining second data to valid data classification by action node；And
Data row point subelement, for sorted second data are carried out row's point.
As the preferred technical solution of the present invention, using Principal Component Analysis algorithms to data
Dimension carries out reduction process.
The data visualization processing method identified based on multidimensional of the present invention, by including step：By Internet technology
The collection client big data related to selected main body；Client big data is imported to into distributed data base, and to importing
Data carry out arrangement classification, while carrying out data screening duplicate removal, and obtain valid data after excluding useless invalid data；To have
Effect data carry out drop latitude process and obtain low-dimensional degrees of data；Using HTML 5<canvas>Label, is called by JavaScript
The API of canvas realizes the image rendering to low latitudes data so that the present invention can be by the data with ten million magnitude by letter
Single visualization datagram displaying is processed, it is achieved thereby that the Data Analysis Services that multidimensional label method is combined with data visualization
Method, and by the big data of ten million magnitude, identify through multidimensional and visualization processes data into simple datagram, can be directly perceived
Ground facilitates the data prediction of available data analysis, data management and later stage.
The data visualization processing system identified based on multidimensional of the present invention, by including：Data acquisition module, for leading to
Cross Internet technology and gather the client big data related to selected main body；Data import modul, for by client big data
Distributed data base is imported to, and the data to importing carry out arrangement classification, while carrying out data screening duplicate removal, and exclude useless
Valid data are obtained after invalid data；Data multidimensional identifies processing module, processes for valid data are carried out drop latitude
To low-dimensional degrees of data；And data visualization processing module, for using HTML 5<canvas>Label, by JavaScript
The API of canvas is called to realize the image rendering to low latitudes data so that the data with ten million magnitude can be led to by the present invention
Cross, realize the Data Analysis Services that multidimensional label method is combined with data visualization
Method, and by the big data of ten million magnitude, identify through multidimensional and visualization processes data into simple datagram, can be directly perceived
Ground facilitates the data prediction of available data analysis, data management and later stage.
Description of the drawings
The present invention is further detailed explanation with reference to the accompanying drawings and detailed description.
The method flow of the embodiment that Fig. 1 is provided based on the data visualization processing method that multidimensional is identified for the present invention
Figure；
Valid data are carried out drop latitude process for the present invention and obtain the method flow diagram that low-dimensional degrees of data is provided by Fig. 2；
Fig. 3 adopts multidimensional label method for the present invention, reduces the method flow diagram that the dimension of valid data is provided；
The structured flowchart of the embodiment that Fig. 4 is provided based on the data visualization processing system that multidimensional is identified for the present invention.
Fig. 5 identifies the structural representation that processing module is provided for data multidimensional of the present invention；
The structural representation that Fig. 6 is provided for dimension-reduction treatment unit of the present invention.
The object of the invention is realized, functional characteristics and advantage will be described further in conjunction with the embodiments referring to the drawings.
Specific embodiment
It should be appreciated that specific embodiment described herein is not intended to limit the present invention only to explain the present invention.
The method flow of the embodiment that Fig. 1 is provided based on the data visualization processing method that multidimensional is identified for the present invention
Figure.
Such as Fig. 1, a kind of data visualization processing method identified based on multidimensional, comprise the following steps：
Step 11, gathers the client big data related to selected main body by Internet technology；
Client big data is imported to distributed data base, and the data to importing carries out arrangement classification, together by step 12
Shi Jinhang data screening duplicate removals, and valid data are obtained after excluding useless invalid data；
Valid data are carried out drop latitude process and obtain low-dimensional degrees of data by step 13；
Step 14, using HTML 5<canvas>Label, calls the API of canvas to realize to low latitude by JavaScript
The image rendering of degrees of data, so that the data with ten million magnitude are processed by the displaying of simple visualization datagram.
In being embodied as, in step 11, client includes smart mobile phone or computer terminal, and client big data exists for user
By the Internet pair APP related to selected main body or the peration data of browser in client, which includes APP or browser
Frequency of usage, using duration, click on situation, and related business process in each node data.
In being embodied as, in step 12 using distributed data library storage client big data, client big data
Collection, its main feature and challenge are that number of concurrent is high because while be possible to have thousands of user conducting interviews and
Operation, data throughout are very big.For example：The Taobao in double 11 periods and 12306 nets in spring transportation period, their concurrent visit capacities
Million magnitudes are reached in peak value, the too high problem of database loads can be solved by way of distributed data base process.
In being embodied as, step 13 specifically includes following steps, as shown in Figure 2：
Step 131, defines target, it is determined that the target closely related with selected theme；
Step 132, using multidimensional label method, reduces the dimension of valid data；
Step 133, with minimum dimension fitting output data in the way of space diagram.
In order to allow those skilled in the art to be best understood from technical scheme, step 131 is exemplified below.
For example：Research user is to the interest level of the advertisement of certain brand each product it is necessary to being selected to describe the one of this feature
Number of variables index, that is, the target for being defined, including such as：Advertisement issue rate, displaying rate, clicking rate, the download rate of product, produce
Installation rate of product etc..
In being embodied as, step 132 specifically includes following steps：
Valid data are divided by step 1321 by different dimensions；
Step 1322, obtains first data to valid data classification by timing node；
Step 1323, obtains second data to valid data classification by action node；
Sorted second data are carried out row's point by step 1324.
In being embodied as, during above-mentioned steps 132 reduce valid data dimension using multidimensional identification method, preferably
Reduction process is carried out to the dimension of data using Principal Component Analysis algorithms.
In being embodied as, the computational methods of Principal Component Analysis algorithms are as follows：
If n-dimensional vector w is a change in coordinate axis direction of target subspace, the change in coordinate axis direction is referred to as map vector, maximum
Change the variance equation below after data mapping：
Wherein m is the number of data instance, and xi is that the vector table of data instance i reaches,It is the average of all data instances
Vector；
It is the matrix comprising all map vectors for column vector to define W, through linear algebraic transformation, can obtain following excellent
Change object function：
W'W=I is that each feature is orthogonal, does not have redundancy between so every dimension.
The wherein mark of tr representing matrixs, A is data covariance matrix.
Optimum W is as column vector structure by k before the data covariance matrix maximum corresponding characteristic vector of eigenvalue
Into, these characteristic vectors form one group of orthogonal basis and best remain the information in data.
Output therein is exactly Y=W'X, is to be reduced to k dimensions by the original dimension of X.
The structured flowchart of the embodiment that Fig. 4 is provided based on the data visualization processing system that multidimensional is identified for the present invention.
As shown in figure 4, a kind of data visualization processing system identified based on multidimensional, including：
Data acquisition module 21, for gathering the client big data related to selected main body by Internet technology；
Data import modul 22, for client big data is imported to distributed data base, and the data to importing are entered
Row is arranged to be sorted out, while carrying out data screening duplicate removal, and obtains valid data after excluding useless invalid data；
Data multidimensional scale processing module 23, obtains low-dimensional degrees of data for valid data are carried out drop latitude process；With
And
Data visualization processing module 24, for using HTML 5<canvas>Label, is called by JavaScript
The API of canvas realizes the image rendering to low latitudes data, so that the data with ten million magnitude are by simple visual
Change datagram displaying to process.
As shown in figure 5, the data multidimensional mark processing module 23 is specifically included：
Target definition unit 231, defines target, it is determined that the target closely related with selected theme；
Dimension-reduction treatment unit 232, using multidimensional label method, reduces the dimension of valid data；And
Dimension determining unit 233, with minimum dimension fitting output data in the way of space diagram.
As shown in fig. 6, dimension-reduction treatment unit 232 is specifically included：
Dimension divides subelement 2321, for dividing to valid data by different dimensions；
Timing node classification subelement 2322, for obtaining first data to valid data classification by timing node；
Action node-classification subelement 2323, for obtaining second data to valid data classification by action node；With
And
Data row point subelement 2324, for sorted second data are carried out row's point.
Preferably, reduction process is carried out to the dimension of data using Principal Component Analysis algorithms.
Although the foregoing describing the specific embodiment of the present invention, those skilled in the art should be appreciated that this
It is merely illustrative of, various changes or modifications can be made to present embodiment, without departing from the principle and essence of the present invention,
Protection scope of the present invention is only limited by the claims that follow.
Claims (9)
1. it is a kind of based on multidimensional identify data visualization processing method, it is characterised in that comprise the following steps：
The client big data related to selected main body is gathered by Internet technology；
Client big data is imported to into distributed data base, and the data to importing carry out arrangement classification, while carrying out data
Screening duplicate removal, and valid data are obtained after excluding useless invalid data；
Valid data are carried out into drop latitude process and obtains low-dimensional degrees of data；
Using HTML 5<canvas>Label, calls the API of canvas to realize the figure to low latitudes data by JavaScript
As drawing, so that the data with ten million magnitude are processed by the displaying of simple visualization datagram.
2. according to the data visualization processing method identified based on multidimensional described in claim 1, it is characterised in that by significant figure
Obtain low-dimensional degrees of data and specifically include following steps according to carrying out dropping latitude processing：
Target is defined, it is determined that the target closely related with selected theme；
Using multidimensional label method, the dimension of valid data is reduced；
With minimum dimension fitting output data in the way of space diagram.
3. according to the data visualization processing method identified based on multidimensional described in claim 2, it is characterised in that adopt multidimensional
Label method, the dimension for reducing valid data specifically include following steps：
Valid data are divided by different dimensions；
First data are obtained to valid data classification by timing node；
Second data are obtained to valid data classification by action node；
Sorted second data are carried out into row's point.
4. according to the data visualization processing method identified based on multidimensional described in claim 3, it is characterised in that adopt
Principal Component Analysis algorithms carry out reduction process to the dimension of data.
5. according to the data visualization processing method identified based on multidimensional described in claim 4, it is characterised in that
The computational methods of Principal Component Analysis algorithms are as follows：
If n-dimensional vector w is a change in coordinate axis direction of target subspace, the change in coordinate axis direction is referred to as map vector, maximizes number
According to the variance equation below after mapping：
Wherein m is the number of data instance, and xi is that the vector table of data instance i reaches, and is the average vector of all data instances；
W is the matrix comprising all map vectors for column vector, through linear algebraic transformation, obtains following optimization object function：
The wherein mark of tr representing matrixs, A is data covariance matrix.
6. it is a kind of based on multidimensional identify data visualization processing system, it is characterised in that include：
Data acquisition module, for gathering the client big data related to selected main body by Internet technology；
Data import modul, for client big data is imported to distributed data base, and the data to importing are arranged
Sort out, while carrying out data screening duplicate removal, and valid data are obtained after excluding useless invalid data；
Data multidimensional identifies processing module, obtains low-dimensional degrees of data for valid data are carried out drop latitude process；And
Data visualization processing module, for using HTML 5<canvas>Label, calls canvas's by JavaScript
API realizes the image rendering to low latitudes data, so that the data with ten million magnitude are by simple visualization datagram
Displaying is processed.
7. according to the data visualization processing method identified based on multidimensional described in claim 6, it is characterised in that the data
Multidimensional mark processing module includes：
Target definition unit, defines target, it is determined that the target closely related with selected theme；
Dimension-reduction treatment unit, using multidimensional label method, reduces the dimension of valid data；And
Dimension determining unit, with minimum dimension fitting output data in the way of space diagram.
8. according to the data visualization processing system identified based on multidimensional described in claim 7, it is characterised in that dimension-reduction treatment
Unit is specifically included：
Dimension divides subelement, for dividing to valid data by different dimensions；
Timing node classification subelement, for obtaining first data to valid data classification by timing node；
Action node-classification subelement, for obtaining second data to valid data classification by action node；And
Data row point subelement, for sorted second data are carried out row's point.
9., according to the data visualization processing system identified based on multidimensional described in any one of claim 6 to 8, its feature exists
In reduction process being carried out to the dimension of data using Principal Component Analysis algorithms.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611181383.2A CN106599234A (en) | 2016-12-20 | 2016-12-20 | Data visualization processing method and system based on multidimensional identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611181383.2A CN106599234A (en) | 2016-12-20 | 2016-12-20 | Data visualization processing method and system based on multidimensional identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106599234A true CN106599234A (en) | 2017-04-26 |
Family
ID=58601826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611181383.2A Pending CN106599234A (en) | 2016-12-20 | 2016-12-20 | Data visualization processing method and system based on multidimensional identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106599234A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107317725A (en) * | 2017-06-14 | 2017-11-03 | 物链（北京）科技有限公司 | The method for visualizing and system of a kind of block chain |
| CN108732534A (en) * | 2018-04-19 | 2018-11-02 | 天津大学 | A kind of multi-tag Cooperative Localization Method based on weighting MDS |
| CN112084179A (en) * | 2020-09-02 | 2020-12-15 | 北京锐安科技有限公司 | Data processing method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103354928A (en) * | 2012-02-03 | 2013-10-16 | 日本电气株式会社 | Device, method, and program for visualization of multi-dimensional data |
| US20140324864A1 (en) * | 2013-04-12 | 2014-10-30 | Objectvideo, Inc. | Graph matching by sub-graph grouping and indexing |
| CN104834716A (en) * | 2015-05-11 | 2015-08-12 | 浪潮集团有限公司 | Dimension-reduced mapped big data visualizing method |
| CN105243469A (en) * | 2015-09-17 | 2016-01-13 | 上海寰信网络信息技术有限公司 | Method for mapping from multidimensional space to low-dimensional space, and display method and system |
| CN106131017A (en) * | 2016-07-14 | 2016-11-16 | 何钟柱 | Cloud computing information security visualization system based on trust computing |
-
2016
- 2016-12-20 CN CN201611181383.2A patent/CN106599234A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103354928A (en) * | 2012-02-03 | 2013-10-16 | 日本电气株式会社 | Device, method, and program for visualization of multi-dimensional data |
| US20140324864A1 (en) * | 2013-04-12 | 2014-10-30 | Objectvideo, Inc. | Graph matching by sub-graph grouping and indexing |
| CN104834716A (en) * | 2015-05-11 | 2015-08-12 | 浪潮集团有限公司 | Dimension-reduced mapped big data visualizing method |
| CN105243469A (en) * | 2015-09-17 | 2016-01-13 | 上海寰信网络信息技术有限公司 | Method for mapping from multidimensional space to low-dimensional space, and display method and system |
| CN106131017A (en) * | 2016-07-14 | 2016-11-16 | 何钟柱 | Cloud computing information security visualization system based on trust computing |
Non-Patent Citations (2)
| Title |
|---|
| 彭非等: "《发展测度论-中国的实践与思考》", 30 April 2014, 中国社会出版社 * |
| 马建堂: "《大数据：政府统计的新机遇》", 31 August 2015, 中国统计出版社 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107317725A (en) * | 2017-06-14 | 2017-11-03 | 物链（北京）科技有限公司 | The method for visualizing and system of a kind of block chain |
| CN108732534A (en) * | 2018-04-19 | 2018-11-02 | 天津大学 | A kind of multi-tag Cooperative Localization Method based on weighting MDS |
| CN112084179A (en) * | 2020-09-02 | 2020-12-15 | 北京锐安科技有限公司 | Data processing method, device, equipment and storage medium |
| CN112084179B (en) * | 2020-09-02 | 2023-11-07 | 北京锐安科技有限公司 | Data processing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Frazier et al. | Landscape metrics: past progress and future directions | |
| Fox et al. | Assessing the accuracy and stability of variable selection methods for random forest modeling in ecology | |
| Aiello‐Lammens et al. | spThin: an R package for spatial thinning of species occurrence records for use in ecological niche models | |
| Perry et al. | A comparison of methods for the statistical analysis of spatial point patterns in plant ecology | |
| Acar et al. | Unsupervised multiway data analysis: A literature survey | |
| Zhao et al. | A 2-D wavelet decomposition-based bag-of-visual-words model for land-use scene classification | |
| US20120047123A1 (en) | System and method for document analysis, processing and information extraction | |
| Abd El Aziz et al. | Automatic detection of galaxy type from datasets of galaxies image based on image retrieval approach | |
| Wang et al. | Incorporation of texture information in a SVM method for classifying salt cedar in western China | |
| CN106599234A (en) | Data visualization processing method and system based on multidimensional identification | |
| Waldner et al. | The impact of training class proportions on binary cropland classification | |
| Al-Sabouni et al. | Vertical niche separation control of diversity and size disparity in planktonic foraminifera | |
| Smale | Monitoring marine macroalgae: the influence of spatial scale on the usefulness of biodiversity surrogates | |
| CN106844588A (en) | A kind of analysis method and system of the user behavior data based on web crawlers | |
| Yu et al. | Salient feature extraction for hyperspectral image classification | |
| Yoshizumi et al. | A review of geospatial content in IEEE visualization publications | |
| Patil et al. | Multiscale advanced raster map analysis system: definition, design and development | |
| De Vries et al. | Parallel streaming signature em-tree: A clustering algorithm for web scale applications | |
| Guo et al. | Saliency detection on sampled images for tag ranking | |
| Mao et al. | Time-varying graph signals reconstruction | |
| Moumtzidou et al. | Discovery of environmental resources based on heatmap recognition | |
| Ma et al. | Modeling bird species richness at multiple spatial scales using two-dimensional wavelet analysis | |
| Sahoo et al. | On the study of GRBF and polynomial kernel based support vector machine in web logs | |
| Dick et al. | Gas prices of america: The machine-augmented crowd-sourcing era | |
| Liu et al. | Sequential pattern mining of land cover dynamics based on time-series remote sensing images |
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: 20170426 |
|
| RJ01 | Rejection of invention patent application after publication |