CN106294298B - Data visualization analysis method based on the sequence of more attributes and application - Google Patents
Data visualization analysis method based on the sequence of more attributes and application Download PDFInfo
- Publication number
- CN106294298B CN106294298B CN201610601476.XA CN201610601476A CN106294298B CN 106294298 B CN106294298 B CN 106294298B CN 201610601476 A CN201610601476 A CN 201610601476A CN 106294298 B CN106294298 B CN 106294298B
- Authority
- CN
- China
- Prior art keywords
- data
- attribute
- attributes
- value
- ranking
- 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
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/174—Form filling; Merging
-
- 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
- G06F16/24—Querying
- G06F16/248—Presentation of query results
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of data visualization analysis method based on the sequence of more attributes and applications, in conjunction with parallel coordinate visualization method and bar chart, numerical value is expressed with bar chart length, high dimensional data is shown with parallel coordinates, overall ranking is carried out to more attributes, realizes while reflecting the incidence relation between the ranking of data and data different attribute.Including:Obtain the attribute value of each risks and assumptions in data;Weights are assigned to attribute value and generate comprehensive evaluation value;Attribute value is shown by way of bar chart, is classified to comprehensive evaluation value, comprehensive assessment risks and assumptions are extracted, is shown by radiating ring;And all risk factor is connected using parallel coordinates;Using Sang Jitu and map denotation;Property parameters, which are changed, by interactive operation shows more attribute ranking results.The method of the present invention can carry out more attribute ranking visual analysis towards pesticide residue data, the integrated information data of university, omics data etc..
Description
Technical field
The invention belongs to information visualization fields, are related to data visualization analysis method, more particularly to a kind of based on more attributes
The data visualization analysis method of sequence and application.
Background technology
Sequence is a kind of common data analysing method, is widely used in every field.Sequence is in our daily life
In be seen everywhere, whether national university's sequence or profession income sort.Sequence is ubiquitous.The critical function of sequence
It is to aid in us and guidance is provided when browsing thing, such as what is considered as " good ", " prevalence ", " high quality ".It can expire
Sufficient people's needs to filter some contents to obtain information.
Sequence can be the sequence of single attribute, i.e., the value based on single attribute is ranked up, such as best-seller list, according to book
The sales volume of nationality carrys out ranking;Can also be more attribute sequences, i.e., the integrated value based on multiple attributes is ranked up, such as basis
Price, per gallon gasoline mileage number, appearance etc. determine the ranking of automobile;Also with good grounds level of education, publish thesis feelings
The factors such as condition, prize-winning situation, the horizontal influence power of teaching and administrative staff carry out World University's ranking;According to urban population quantity, GDP, city
Road construction, greening situation come ranking most city good for habitation etc..To a large amount of sequence involved in the data analysis of field of food safety
Analysis, such as in the analysis to Pesticide Residues testing result, not only will to pesticide detection species number, pesticide detected level,
Multiple independent single attributes such as the exceeded frequency of pesticide frequency, pesticide, banning drugs frequency are ranked up analysis;It also needs to pair
The pesticide residual contamination degree composite index of each department, the security index of various agricultural product are ranked up analysis, i.e., to belong to more
Property sequence;Sometimes it also needs to the association between each single attribute and its influence to integrated ordered result is analyzed.
From method, single attribute sequence research lays particular emphasis on the sort algorithm of precise and high efficiency, such as bubble sort, selection row
Sequence, quicksort etc.;And more attribute sequences then need first to be melted the independent single attribute of multiple isomeries according to model or expertise
It is combined into a synthesized attribute, is then ranked up again with sort algorithm.How the attribute of multiple independent isomeries to be permeated
A synthesized attribute is simultaneously ranked up, and is analyzed the association between each independent attribute and its is one to integrated ordered influence and has much
The problem of challenge.The phenomenon implied in data and rule is presented by visualized graphs in the Information Visualization Technology occurred in recent years
Rule, to solve such issues that, provide new tool.
More attribute sequence method for visualizing include mainly based on point, based on line and based on face three classes.Wherein, the side based on point
Method visualizes ranking result using position as visual variable.Such as scatter plot (scatterplots) and scatterplot matrix
(scatterplot matrix) can be used to compare the ranking of two things.In document [1] (A.P.Sawant and
C.G.Healey.Visualizing multidimensional query results using animation[J].In
Proc.IS&T/SPIE Electronic Imaging.2008:680-904) in, Sawant and Healey utilize hyperspace
Spiral is filled to show film rating result.Each film is indicated with the different bar shaped column of height, is placed in a spiral.Position
The evaluation of mapping film from inside to outside is from high to low.In method based on line, line can be used for connecting multiple values or more multiple
Ranking results, such as slope figure (Slop graph) and parallel coordinates (Parallel Coordinates).In document [2]
(M.Batty.Rank clocks[J].Nature.2006,444:592-596.) in, Batty devises Rank Clocks, profit
With radar map, a kind of special parallel coordinates shows city of the world's population sequence variation across several centuries.In document [3]
(Hui Lei,Jing Xia,FanzhouGuo.Visual Exploration of Latent Ranking Evolutions
in Time Series[J].Journal of Visualization,2016:Pp 1-13) in, RankEvo methods are using flat
Analysis time sequence of row coordinate coming potentially sorts evolution.Document [4] (Min Lu, ZuchaoWang,
XiaoruYuan.TrajRank Exploring Travel Behaviour on a Route by Trajectory
Ranking[C].In Proceedings of IEEE Pacific VisualizationSymposium.2015:311-
318) in, car data is hired out in the application of TrajRank methods, investigates its running orbit, studies the path of vehicle traveling and to each section
Track sequence.Document [5] (David H.S.Chung, Matthew L.Parry, Iwan
W.Griffiths.Knowledge-Assisted Ranking:A Visual Analytic Application for
Sport Event Data [J] .IEEE Computer Graphics.2015) record Knowledge-Assisted
Ranking is the tool of an analysis sports data, which can be involved in multiple attributes, criteria for classification,
The sports video data often changed is ranked up.In method based on face, area is that another is used for encoding quantitative data
Effective visible sensation variable, such as bar chart and stacking are schemed.Wherein theme river (Themeriver) just belongs to one kind of stacking figure.Document
[6](Conglei Shi,Weiwei Cui,ShixiaLiu.RankExplorer Visualization of Ranking
Changes in Large Time Series Data[J].IEEE Transactions on Visualization and
Computer Graphics,2012,18(12):2669-2678.) record RankExplorer by expand Themeriver,
Data are divided into graded category, with the variation of symbology grade.In document [7] (Michael Behrisch, James
Davey,Svenja Simon,et al.Visual comparison of orderings and rankings[M].In
Proceedings of theEuroVis Workshop on Visual Analytics.2013) in, Behrisch uses one
A radial direction node-link indicates that different sequences, this method depend on symbol, each small doughnut to be used for comparing a pair of of row
Sequence.
Nowadays, the visualization technique of single view cannot meet the demand of growth and the analysis of data, a variety of views
In conjunction with being observed that more information.In document [8] (J.Seo and B.Shneiderman.A rank-by-feature
framework for unsupervised multidimensional data exploration using low
dimensionalprojections[J].In Proc.IEEE InfoVis.2004:65-72) in, Seo proposes rank-by-
Feature methods, to predict cube with helping custom system, with orderly block diagram to the ranking results of single attribute
Displaying, with two attribute ranking results displayings of scatter plot pair.In addition, there are also visualization techniques by line chart and histogram knot
Close, so as to can display properties occurrence and indicate the association between attribute.Document [9] (Saori Okubo,
TomoyaIwakura,KazuoMisue.Trend Analysis Tool with Simultaneous Visualization
of Rank and Value.17th International Conference on Information Visualisation
[C] .2013,517-522.) described in trend analysis tool can show sequence and the value of event simultaneously at any time, with text
This label represents event category, indicates to sort with color saturation;Simultaneously with the change of the ribbon mark event ordering of corresponding color
Change.In document [10] (Samuel Gratzl, AlexanderLex, NilsGehlenborg, et al.LineUp:Visual
Analysis of Multi-Attribute Rankings[J].IEEE Transactions on Visualization
and Computer Graphics,2013,19(12):In 2277-2286), LineUp approach application histograms it is expansible
Property, main advantage is can interactively to improve weights and attribute mapping, is easy to track the ability of attribute sequence.
Existing method for visualizing can not often take into account more attributes and single attribute sequence while show when solution is sorted
Show;And the method for visualizing that individually attribute is ranked up, it often can not intuitively show its attribute value.In addition without being directed to agriculture
The sequence visual analysis method of medicine residual data.
Invention content
In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a kind of data visualization analysis sorted based on more attributes
Bar chart length is expressed the thought of numerical value by methods and applications by a kind of method for visualizing of parallel coordinates combination bar chart
Show that the method for visualizing of high dimensional data is combined with parallel coordinates, and application overall ranking algorithm carries out comprehensive row to more attributes
Name, this method can reflect the incidence relation between the ranking of data and different attribute simultaneously.Can towards pesticide residue data,
The integrated information data of university, omics data etc. carry out more attribute ranking visual analysis.
Technical solution provided by the invention is:
A kind of data visualization analysis method based on the sequence of more attributes, by combining parallel coordinate visualization method and bar shaped
Figure method for visualizing, numerical value is expressed with bar chart length, shows high dimensional data with parallel coordinates, and pass through overall ranking calculating side
Method carries out overall ranking to more attributes, realizes while reflecting that the association between the ranking of data and data different attribute is closed
System;Include the following steps:
Step 1: it is for statistical analysis to initial data, obtain the attribute value of each risks and assumptions in data;
Step 2: the attribute value to the risks and assumptions obtained in step 1 assigns weights, comprehensive evaluation model is established, is generated comprehensive
Close evaluation of estimate;
Step 3: to the attribute values of the risks and assumptions obtained in step 1 by way of bar chart sequencing display, to step
The comprehensive evaluation value obtained in rapid two is classified, and the risks and assumptions for comprehensive assessment are extracted, and is shown by radiating ring;And
A kind of all risk factor of things is connected using parallel coordinates;
Step 4: using Sang Jitu and map, to show more data informations, visualization result is obtained;
Step 5: operation can be interacted by interface, property parameters is changed by interactive operation, the sequence of more attributes is tied
Fruit is shown, for being analyzed and being compared.
For the above-mentioned data visualization analysis method based on the sequence of more attributes, further, knot is visualized described in step 4
Fruit includes:The visualization interface leftmost side is the data screening frame for interacting operation;It is topmost sequence constitutional diagram, including
One parallel coordinates adds bar chart and radiates the constitutional diagram of ring;The lower left corner is thermal map;The lower right corner is radiation ring enlarged drawing;The most right side
It is Sang Jitu.
For the above-mentioned data visualization analysis method based on the sequence of more attributes, further, step 2 is assessed using expert
Method assigns weights.
For the above-mentioned data visualization analysis method based on the sequence of more attributes, further, in embodiments of the present invention, institute
It is pesticide residue data to state initial data, and the risks and assumptions include that agricultural product middle peasant's medicine inspection goes out kind and the frequency, pesticide are exceeded
The kind and frequency, the high highly toxic pesticide detection kind and frequency.
For the above-mentioned data visualization analysis method based on the sequence of more attributes, further, the method is applied to pesticide
The more attribute sequence visual analysis of residual data, the compositive university ranking analysis or disciplines ranking analysis and assessment.
The above-mentioned data visualization analysis method based on the sequence of more attributes is applied to the more attribute sequences of pesticide residue data can
Depending on analysis, wherein the risks and assumptions include that agricultural product middle peasant's medicine inspection goes out kind and the exceeded kind of the frequency, pesticide and the frequency, height
Highly toxic pesticide detects kind and the frequency;The attribute value of the risks and assumptions is obtained by classified statistic method;Pass through expert estimation
Method is that weight is arranged in each attribute;By establishing agricultural product integrated pollution evaluation model, according to the attribute of each agricultural product into
Row weighted score obtains comprehensive evaluation value;Obtained attribute value is ranked up according to comprehensive evaluation value;By combining parallel seat
Method for visualizing and bar chart method for visualizing are marked, numerical value is expressed with bar chart length, high dimensional data is shown with parallel coordinates, and
Overall ranking is carried out to more attributes by overall ranking computational methods, realizes while reflecting the ranking and data difference of data
Incidence relation between attribute.
In the application of the above-mentioned data visualization analysis method based on the sequence of more attributes, further, the agricultural product synthesis
Pollution evaluation model is specially formula 1:
In formula 1:
αi∈ A={ α1,α2…αnIndicate a kind of agricultural product;
wi∈ W={ w1,w2…wnIndicate the corresponding weight of attribute;
pi∈ P={ p1,p2…pnIndicate the value of each attribute;
pi' indicate the attribute value after normalization;
Attribute vector P is mapped to the range of 0-1, is normalized multiple attributes by formula 2:
Wherein, pi* the result after attribute i normalization is indicated;piResult before being attribute i normalization;pminAnd pmaxRespectively
It is the minimum value and maximum value in multiple attributes.
Compared with prior art, the beneficial effects of the invention are as follows:
The present invention proposes a kind of data visualization analysis method to sort based on more attributes and application, passes through a kind of parallel coordinates
In conjunction with the method for visualizing of bar chart, bar chart length is expressed into the thought of numerical value and parallel coordinates shows the visual of high dimensional data
Change method is combined, and application overall ranking algorithm carries out overall ranking to more attributes, and this method can reflect data simultaneously
Incidence relation between ranking and different attribute.It can be towards pesticide residue data, the integrated information data of university, omics data
Deng the more attribute ranking visual analysis of progress.More attributes sequence visual analysis method towards pesticide residue data can show agriculture simultaneously
Incidence relation between single attribute ranking of medicine residual data and the more attribute rankings and attribute of synthesis helps user to check data
Ranking Long-term change trend contributes to further data analysis.
Description of the drawings
Fig. 1 is the method flow frame that the embodiment of the present invention carries out more attributes sequence visual analysis for the residual detection data of agriculture
Figure.
Fig. 2 is the sectional drawing of the visual layout using parallel coordinates+bar chart in the embodiment of the present invention;
Wherein, downward arrow r indicates ranking from high to low;The wide v of each rectangular block indicates attribute value;On the left of each axis
Number 1,2,3... indicate specific ranking;A, b, c indicate three attributes respectively;D indicates to decline;U indicates to rise.
Fig. 3 is that the pesticide based on radiation ring in the embodiment of the present invention detects distribution of results figure;
Wherein, inner ring 1,2,3, n indicate parent attribute value accounting, corresponding 1.1,1.2,1.3 be three of parent attribute 1
The sub- corresponding accounting of attribute value;φ0Indicate the initial angle of radiation inner ring, secondly angle is φ1、φ2、φn;Curved arrow table
Show that it is counterclockwise to draw fan-shaped direction.
Fig. 4 is the sectional drawing of more attribute rankings visual layout towards pesticide residue data in the embodiment of the present invention;
Wherein, 1,2,3... indicate the ranking of each attribute.
Fig. 5 is the Sang Jitu sectional drawings that middle peasant's medicine inspection of the embodiment of the present invention goes out that pesticide variety, the frequency and exceeded situation are distributed.
Fig. 6 is that the pesticide residue ranking of more attribute ranking view combination Sang Jitu and map visualizes in the embodiment of the present invention
As a result sectional drawing.
Specific implementation mode
Below in conjunction with the accompanying drawings, the present invention, the model of but do not limit the invention in any way are further described by embodiment
It encloses.
The present invention provides a kind of data visualization analysis method to sort based on more attributes and application, passes through a kind of parallel coordinates
In conjunction with the method for visualizing of bar chart, bar chart length is expressed into the thought of numerical value and parallel coordinates shows the visual of high dimensional data
Change method is combined, and application overall ranking algorithm carries out overall ranking to more attributes, and this method can reflect data simultaneously
Incidence relation between ranking and different attribute.It can be towards pesticide residue data, the integrated information data of university, omics data
Deng the more attribute ranking visual analysis of progress.
Data visualization method provided by the invention for ranking problem needs the ranking and occurrence of expressing attribute.
The concrete operation step of this method is:
Step 1: it is for statistical analysis to initial data, obtain the attribute value of each risks and assumptions;
In specific implementation, risks and assumptions are determined by the requirement documents of relevant departments;For example, being directed to pesticide residue
Data, risks and assumptions include that agricultural product middle peasant's medicine inspection goes out kind and the exceeded kind of the frequency, pesticide and the frequency, the detection of high highly toxic pesticide
Kind and the frequency, these data are by being obtained to the statistic of classification of part initial data.As initial data is examined comprising agricultural product
Each pesticide volume gone out, by with Chinese pesticide residue maximum limit the quantity MRL compared, be more than MRL indicate it is exceeded, be less than be exactly
It is not exceeded.Exceeded pesticide species and the frequency are counted with this.
Step 2: the attribute value to the risks and assumptions obtained in step 1 assigns weights, comprehensive evaluation value is generated;
Expert's Evaluation Method can be used and assign weights, expert appraisal approach is on the basis of qualitatively and quantitatively analyzing, with marking etc.
Mode makes quantitative assessment.Each attribute value is first normalized to a number in a 0-1, is multiplied by corresponding weights, is owned
It is 1 that the weights of attribute, which mutually sum it up, and the determination of weights is to be provided by field multidigit expert and be averaged determination.Finally by all categories
Property value be multiplied by the number adduction after weights and obtain comprehensive evaluation value.
Step 3: to the attribute value that is obtained in step 1 by way of bar chart sequencing display, to being obtained in step 1
The attribute values of risks and assumptions classify, extract the risks and assumptions for comprehensive assessment, shown by radiating ring, and make to put down
Row coordinate connects a kind of all risk factor of things (for example, agricultural product in pesticide residue data).
Step 4: using Sang Jitu and map, to show more data informations;
Step 5: user can change property parameters repeatedly, observation ranking becomes by interface alternation function in data screening frame
Change.
By the operation of above-mentioned steps, final visualization result is obtained.In the embodiment of the present invention, final visualization result is adopted
With following displaying:The leftmost side is data screening frame;It is topmost the master map of sequence, including a parallel coordinates adds and bar chart and puts
Penetrate the constitutional diagram of ring;The lower left corner is thermal map;The lower right corner is the enlarged drawing of a radiation ring;The most right side is Sang Jitu.
Following embodiment is directed to national Practice for Pesticide Residue in Agricultural Products testing result data, and what is provided through the invention is belonged to more based on
Property sequence data visualization analysis method be ranked up and visualize.Fig. 1 be the embodiment of the present invention for the residual detection data of agriculture into
The method flow block diagram of the more attribute sequence visual analysis of row.Initial data is as shown in table 1:
1 Practice for Pesticide Residue in Agricultural Products initial data of table
Root, according to the attribute and weights of setting overall merit, establishes Multi-attribute synthetic evaluation model (the present embodiment in agriculture residue
Establish agricultural product integrated pollution evaluation model):
More attribute sequencing problems are a "black box" problems, since we cannot accurately describe the relationship between them,
So weight that cannot clearly between distributive property.The overall merit of each things is obtained by Multi-attribute synthetic evaluation model
Value simultaneously is used to sort.
Multiattribute comprehensive assessment uses fairly simple intuitive expert appraisal approach in the method, asks n (n>5) position neck
Weight is arranged for each attribute in a manner of marking in domain expert, and the scoring of an attribute of each expert couple is s1, s2...sn.It should
The weight of attribute is (s1+s2+...+sn)/n, and the attribute weight score further according to each agricultural product obtains total score S.Thus this hair
One agricultural product integrated pollution evaluation model of bright proposition, the attribute weight total score S of agricultural product is specifically calculated by formula 1, i.e.,
Multiattribute assessment model:
In formula 1:
αi∈ A={ α1,α2…αnIndicate a kind of agricultural product;
wi∈ W={ w1,w2…wnIndicate the corresponding weight of attribute;
pi∈ P={ p1,p2…pnIndicate the value of each attribute;
pi' indicate the attribute value after normalization;Due to the isomerism of attribute, unit and the value range difference of attribute need
It normalizes, attribute vector P is mapped to the range of 0-1.By taking tomato as an example, eight attributes are normalized by formula 2 first:
Wherein, pi* the result after attribute i normalization is indicated;piResult before being attribute i normalization;pminAnd pmaxRespectively
It is the minimum value and maximum value in eight attributes.
piMultiplied by with respective weights, S (tomato)=0.111 is obtained.
Using ranking method for visualizing proposed by the present invention, visualized operation, concrete operations step are carried out to above-mentioned data
Suddenly it is:
Step 1: it is for statistical analysis to initial data (table 1), obtain the attribute value of each risks and assumptions, including agricultural production
Product middle peasant's medicine inspection goes out kind, the frequency;The exceeded kind of pesticide, the frequency;High highly toxic pesticide detection kind, the frequency, banning drugs detection kind,
The frequency, such as table 2.
2 Practice for Pesticide Residue in Agricultural Products risks and assumptions statistical data of table
Step 2: carrying out descending sort to the attribute value obtained in step 1;
Step 3: obtain weights by expert estimation averaging to the association attributes in step 1, wherein overall merit
Each risks and assumptions are obtained according to the classification of the risks and assumptions of 2 Practice for Pesticide Residue in Agricultural Products of table, due to according to correlation analysis, passing through
Correlation coefficient r is calculated in formula 3:
In formula 3, X, Y are respectively two and need the risks and assumptions compared.
Pesticide detects species number and frequency number related coefficient 0.8861, and exceeded kind and the frequency are 0.7195, detects product
Kind and exceeded kind are 0.7501, belong to have significantly correlated.So being directed to this four attributes, we only consider to detect agriculture
Influence of the drug kind number to agricultural product integrated pollution evaluation;Final integrated pollution evaluation detects kind number by pesticide, detects pesticide
Three kinds of factors of the exceeded situation of the frequency and pesticide of toxicity determine.Weights distribution, the power that expert provides are carried out to three kinds of factors
Value distribution such as table 3.Using expert's Evaluation Method, comprehensive evaluation value is generated by formula 1, is sorted according to comprehensive evaluation value, such as table 4.
Each risks and assumptions weights distribution of 3. overall merit of table
4. comprehensive evaluation value ranking of table
Step 4: the attribute value that step 2 is obtained sequencing display by way of bar chart, and connected using parallel coordinates
A kind of all risk factor of agricultural product is connect, as shown in Fig. 2, the length of bar chart indicates that the size of attribute value, each axis indicate
Each attribute, each axis are to arrange in descending order, and the line of a same color connects whole attributes of an agricultural product, passes through sight
The dipping and heaving for examining line can determine whether the variation of sequence.
Step 5: evaluate the aggregation of data of step 3 the attribute value of each risks and assumptions, i.e. the attribute of table 3, with radiation ring
Form show that radiation ring shares three layers, the roundlet of innermost layer is mapped the size of an attribute value by the depth of color.Outside two
Layer annulus, inner and outer ring respectively indicate that an attribute, two attribute are hierarchical relationships.A class of every piece of Regional Representative attribute in ring
Not, it is indicated with different colors.Attribute value is mapped on radiation ring, in mapping process, successively from the center of circle right direction inverse time
Needle is as section prime direction, and the attribute value of the bigger Interval Maps of area is higher, and the phase angle shared by each section is based on
The ratio of attribute value calculates, and i-th layer of angle computation method is as shown in formula 4:
Assuming that i+1 layer, shares j node, i.e. node TijFather node be Ti, then node TijAngle is:
Wherein, it is specific object value, first 0 degree fan-shaped of start angle, statistics of attributes that n, which is the other number .x of Attribute class,
Amount mapping such as Fig. 3.
Step 6: it is used to ask the risks and assumptions radiation ring of overall merit to show in each agricultural product in step 5, according to
The overall ranking sequence of table 4 is shown in last row of parallel coordinates, and is connected with each attribute of the agricultural product before, such as
Fig. 4;
Step 7: the frequency that the pesticide name of 1 agricultural product of table detection and corresponding table 2 are counted and the exceeded frequency
It is shown in the form of Sang Jitu, such as Fig. 5;
It is shown on map in a manner of thermal map Step 8: counting the frequency of whole agricultural product of a provinces and cities,
Such as the thermal map of the lower left corners Fig. 6 figure.
By the operation of above-mentioned steps, final visualization result is obtained, as shown in Figure 6.
The method for visualizing of the present invention is also applied to Detecting Pesticide assessment.Include in Monitoring Pesticide Residues data
Many information, such as pesticide detection kind, the frequency;The exceeded kind of pesticide, the frequency;High highly toxic pesticide detection kind, frequency etc..User
It is often desirable to integrate the pollution degree evaluation that required multiple pesticide residue attributes obtain agricultural product synthesis.To sentence
Determine the quality of different sources agricultural product and the pollution condition of areal difference agricultural product.
In addition, the method for visualizing of the present invention can also be applied in University Rank or disciplinary assessment.The comprehensive row of university
Name index includes school's influence power, quality of instruction, the system of qualifying for teachers, employment rate, paper and publication situation etc..It also belongs to more
Attribute sequencing problem.Each index can be shown with the present invention and obtained overall ranking using comprehensive evaluation model and shown.It learns
Section's assessment is the assessment to the every subjects of a school, such as department of computer science, department of mathematics, Arts Department, management system.Index includes
Ranks of teachers and resource, scientific research level, Talent-cultivating Quality, discipline reputation and discipline etc..Soliciting constituent parts and various aspects extensively
The weight of each index is determined after expert opinion.More attribute sequencing problems are also belonged to, can be indicated with the method for visualizing of the present invention.
It should be noted that the purpose for publicizing and implementing example is to help to further understand the present invention, but the skill of this field
Art personnel are appreciated that:It is not departing from the present invention and spirit and scope of the appended claims, various substitutions and modifications are all
It is possible.Therefore, the present invention should not be limited to embodiment disclosure of that, and the scope of protection of present invention is with claim
Subject to the range that book defines.
Claims (5)
1. a kind of data visualization analysis method based on the sequence of more attributes, by combining parallel coordinate visualization method and bar chart
Method for visualizing, numerical value is expressed with bar chart length, shows high dimensional data with parallel coordinates, and pass through overall ranking computational methods
Overall ranking is carried out to more attributes, realizes while reflecting the incidence relation between the ranking of data and data different attribute;
Include the following steps:
Step 1: it is for statistical analysis to initial data, obtain the attribute value of each risks and assumptions in data;The initial data
For pesticide residue data;
Step 2: the attribute value to the risks and assumptions obtained in step 1 assigns weights, by establishing agricultural product integrated pollution evaluation
Model is weighted score according to the attribute of each agricultural product, generates comprehensive evaluation value;The agricultural product integrated pollution evaluation mould
Type is specially formula 1:
In formula 1:
αi∈ A={ α1,α2...αnIndicate a kind of agricultural product;
wi∈ W={ w1,w2...wnIndicate the corresponding weight of attribute;
pi∈ P={ p1,p2…pnIndicate the value of each attribute;
pi' indicate the attribute value after normalization;
Attribute vector P is mapped to the range of 0-1, is normalized multiple attributes by formula 2:
Wherein, pminAnd pmaxIt is the minimum value and maximum value in multiple attributes respectively;
Step 3: the attribute value to the risks and assumptions obtained in step 1 passes through parallel coordinate visualization according to comprehensive evaluation value
The mode sequencing display of method and bar chart, numerical value is expressed with bar chart length, shows high dimensional data with parallel coordinates, and pass through
Overall ranking computational methods carry out overall ranking to more attributes, realize while reflecting the ranking and data different attribute of data
Between incidence relation;Classify to the comprehensive evaluation value obtained in step 2, extract risk for comprehensive assessment because
Son is shown by radiating ring;And a kind of all risk factor of things is connected using parallel coordinates;Wherein, extraction is for integrating
The risks and assumptions of assessment are obtained according to the classification of the risks and assumptions of Practice for Pesticide Residue in Agricultural Products, and correlation is calculated especially by formula 3
Coefficient r:
In formula 3, X, Y are respectively two and need the risks and assumptions compared;
The risks and assumptions for comprehensive assessment are obtained according to related coefficient;
Step 4: using Sang Jitu and map, for showing more data informations, visualization result is obtained;
Step 5: operation can be interacted by interface, by interactive operation change property parameters to more attribute ranking results into
Row displaying, for being analyzed and being compared.
2. the data visualization analysis method as described in claim 1 based on the sequence of more attributes, characterized in that visual described in step 4
Changing result includes:The visualization interface leftmost side is the data screening frame for interacting operation;It is topmost sequence constitutional diagram,
Add bar chart including a parallel coordinates and radiates the constitutional diagram of ring;The lower left corner is thermal map;The lower right corner is radiation ring enlarged drawing;Most
The right side is Sang Jitu.
3. the data visualization analysis method as described in claim 1 based on the sequence of more attributes, characterized in that step 2 uses expert
Evaluation Method assigns weights.
4. the data visualization analysis method as described in claim 1 based on the sequence of more attributes, characterized in that the risks and assumptions packet
It includes agricultural product middle peasant's medicine inspection and goes out kind and the exceeded kind of the frequency, pesticide and the frequency, the high highly toxic pesticide detection kind and frequency.
5. as described in claim 1 based on more attributes sequence data visualization analysis method, characterized in that step 1 especially by
Classified statistic method obtains the attribute value of the risks and assumptions.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610601476.XA CN106294298B (en) | 2016-07-27 | 2016-07-27 | Data visualization analysis method based on the sequence of more attributes and application |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610601476.XA CN106294298B (en) | 2016-07-27 | 2016-07-27 | Data visualization analysis method based on the sequence of more attributes and application |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106294298A CN106294298A (en) | 2017-01-04 |
CN106294298B true CN106294298B (en) | 2018-11-02 |
Family
ID=57662576
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610601476.XA Active CN106294298B (en) | 2016-07-27 | 2016-07-27 | Data visualization analysis method based on the sequence of more attributes and application |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106294298B (en) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106874407B (en) * | 2017-01-19 | 2020-04-03 | 北京工商大学 | Agricultural residue data fusion processing method and system based on concept hierarchical tree |
CN106920044B (en) * | 2017-03-07 | 2021-01-12 | 北京工商大学 | Pesticide residue risk assessment method based on fuzzy comprehensive evaluation model |
CN107609568A (en) * | 2017-05-26 | 2018-01-19 | 浙江大学 | A kind of method for the influence factor for analyzing ranking trend and ranking stability |
CN109947830B (en) * | 2017-10-19 | 2022-04-26 | 北京京东尚科信息技术有限公司 | Method and apparatus for outputting information |
CN109271567B (en) * | 2018-08-01 | 2022-07-26 | 浙江工业大学 | Multivariable visual analysis method for full-range data |
CN109271464B (en) * | 2018-08-20 | 2021-06-08 | 浙江大学 | Visual ranking system for integrated spatial context |
CN109634699B (en) * | 2018-11-23 | 2022-04-05 | 中国人民解放军63789部队 | Spacecraft abnormal information clustering visualization and interaction method based on mulberry-based map |
CN109460476B (en) * | 2018-12-14 | 2020-10-16 | 东软集团股份有限公司 | Method and device for generating mulberry basic diagram, storage medium and electronic equipment |
CN109784688A (en) * | 2018-12-28 | 2019-05-21 | 上海软科教育信息咨询有限公司 | Level of education visual evaluating method, device, equipment and storage medium |
CN109858827A (en) * | 2019-02-21 | 2019-06-07 | 上海软科教育信息咨询有限公司 | Visualization comparative approach, system, electric terminal and the storage medium of subject level |
CN110631905B (en) * | 2019-08-16 | 2022-07-26 | 中信金属股份有限公司 | Normalization processing method and processing system for automobile body common materials |
CN112632145A (en) * | 2019-09-24 | 2021-04-09 | 北京国双科技有限公司 | Method and device for evaluating new oil and gas reservoir, storage medium and equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101794433A (en) * | 2010-03-03 | 2010-08-04 | 北京师范大学 | City planning implemented ecological security post-evaluation method |
US8346682B2 (en) * | 2009-01-23 | 2013-01-01 | The United States Of America, As Represented By The Secretary Of The Navy | Information assisted visual interface, system, and method for identifying and quantifying multivariate associations |
CN103049580A (en) * | 2013-01-17 | 2013-04-17 | 北京工商大学 | Method and device for visualization of layering data |
CN105303468A (en) * | 2015-11-20 | 2016-02-03 | 国网天津市电力公司 | Comprehensive evaluation method of smart power grid construction based on principal component cluster analysis |
-
2016
- 2016-07-27 CN CN201610601476.XA patent/CN106294298B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8346682B2 (en) * | 2009-01-23 | 2013-01-01 | The United States Of America, As Represented By The Secretary Of The Navy | Information assisted visual interface, system, and method for identifying and quantifying multivariate associations |
CN101794433A (en) * | 2010-03-03 | 2010-08-04 | 北京师范大学 | City planning implemented ecological security post-evaluation method |
CN103049580A (en) * | 2013-01-17 | 2013-04-17 | 北京工商大学 | Method and device for visualization of layering data |
CN105303468A (en) * | 2015-11-20 | 2016-02-03 | 国网天津市电力公司 | Comprehensive evaluation method of smart power grid construction based on principal component cluster analysis |
Non-Patent Citations (1)
Title |
---|
农残检测中多维数据可视分析关键技术研究;蔡进峰;《中国优秀硕士学位论文全文数据库》;20140615(第6期);第37页、第50-51页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106294298A (en) | 2017-01-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106294298B (en) | Data visualization analysis method based on the sequence of more attributes and application | |
Andrienko et al. | Visual analytics of movement: An overview of methods, tools and procedures | |
Najmi et al. | Reviewing the transport domain: An evolutionary bibliometrics and network analysis | |
Labib et al. | Scale effects in remotely sensed greenspace metrics and how to mitigate them for environmental health exposure assessment | |
Carley et al. | Visualization of disciplinary profiles: Enhanced science overlay maps | |
Zhang et al. | A bibliometric review of a decade of research: Big data in business research–Setting a research agenda | |
Gilbert et al. | Communities and hierarchical structures in dynamic social networks: analysis and visualization | |
Sarlin | Macroprudential oversight, risk communication and visualization | |
Andrienko et al. | State transition graphs for semantic analysis of movement behaviours | |
Baier Fuentes et al. | A bibliometric overview of the international journal of interactive multimedia and artificial intelligence | |
Tang et al. | Visual analysis of traffic data based on topic modeling (ChinaVis 2017) | |
Qiang et al. | Multi-scale analysis of linear data in a two-dimensional space | |
Cakmak et al. | Multiscale snapshots: Visual analysis of temporal summaries in dynamic graphs | |
Olawoyin et al. | Privacy-preserving publishing and visualization of spatial-temporal information | |
Andrienko et al. | Visual analytics methods for movement data | |
CN111930957A (en) | Method and apparatus for analyzing intimacy between entities, electronic device, and storage medium | |
Cyriac et al. | A bibliometric review of publication trends in the application of landscape metrics in urban and regional planning | |
Shi et al. | Visual exploration of mobility dynamics based on multi-source mobility datasets and POI information | |
Feranec et al. | Analysis and expert assessment of the semantic similarity between land cover classes | |
Persson | A survey of methods for visualizing spatio-temporal data | |
CN111985576A (en) | Shop address selection method based on decision tree | |
Laroche et al. | Pesticide concentration monitoring: Investigating spatio‐temporal patterns in left censored data | |
Atzmueller et al. | Exploratory subgroup analytics on ubiquitous data | |
Burnicki | Spatio-temporal errors in land–cover change analysis: Implications for accuracy assessment | |
Barb et al. | Applications of PathFinder Network Scaling for improving the ranking of satellite images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB03 | Change of inventor or designer information | ||
CB03 | Change of inventor or designer information |
Inventor after: Chen Yi Inventor after: Tian Shuai Inventor after: Liu Ying Inventor after: Liu Ruijun Inventor after: Wang Yu Inventor before: Chen Yi Inventor before: Liu Ying Inventor before: Liu Ruijun Inventor before: Wang Yu |
|
GR01 | Patent grant | ||
GR01 | Patent grant |