当前位置：网站首页>Data visualization chart summary (I)

Dot matrix diagram (Dot Matrix Chart) Display discrete data in points , The dots of each color represent a specific category , And put them together in matrix form .

It can be used to quickly view the distribution and proportion of different categories in a dataset , And compare it with other data sets , It's easier to find patterns .

Two 、 Dot graph

Dot graph (Point & Figure Charts) Also known as 「P&F chart 」, Use by X and O A series of squares composed of symbols to show the supply and demand relationship of a specific asset .

This chart has nothing to do with time , Mainly focus on the filter price performance of assets ; It also doesn't show the volume of transactions , The purpose is just to show any change in supply and demand , be called 「 breakthrough 」(breakouts).

3、 ... and 、 Arc chart

Arc chart (Arc Diagram) It's another way of data expression besides two-dimensional biaxial chart . In the arc graph , Nodes will follow X Axis placement , Then the arc is used to represent the connection between nodes .

The arc chart is suitable for finding situations where data are common . But the disadvantage is ： It can't show the structure and connection between nodes as clearly as other two axis charts , And too many connections can make the chart difficult to read .

Four 、 Broken line diagram

A line chart is used to display quantitative values over successive intervals or time spans , Most commonly used to show trends and relationships .

Besides , A line chart can also show... In a certain period of time 「 Overall overview 」, Take a look at the development of data over this period of time .

5、 ... and 、 Parallel graph

Parallel graph (Parallel Coordinates Plots) It can display multivariable numerical data , Most suitable for comparing multiple variables at the same time , And show the relationship between them .

When data is dense , Parallel graphs tend to be confusing 、 Illegible . The solution is through interactive technology , Highlight the selected line or lines , And fade all the other lines , Let's focus more on what we're interested in , And filter out the interference data .

6、 ... and 、 Network diagram

Also known as 「 Web map 」 or 「 Node link graph 」, Used to show the type of relationship between things .

These nodes are usually dots or small circles , But you can also use icons .

The main network diagrams are as follows 「 Not oriented 」 and 「 directional 」 Two kinds of . An undirected network diagram shows only the connections between entities , The directed network diagram shows whether the connection is unidirectional or bidirectional （ Through the little arrow ）.

Network graph data capacity is limited , And when there are too many nodes, it will form a similar 「 Hairball 」 The design of , Make it difficult to read .

7、 ... and 、 Pictogram

Pictogram (Pictogram Chart) Also known as 「 Pictograph 」, Use patterns to show the amount of data .

Using patterns can overcome language 、 Cultural and educational differences , It is a more representative method of data display . for instance , If the data is 「5 Vehicles 」, The picture will show 5 A car design .

8、 ... and 、 Histogram

Histograms are suitable for displaying the distribution of data at successive intervals or within a specific period of time , It helps to estimate the location of the value set 、 The upper and lower limits and the determination of whether there are gaps or outliers ; The probability distribution can also be shown roughly .

Nine 、 Density map

Density map (Density Plot) Also known as 「 Density curve 」, Used to display the distribution of data in a continuous period of time .

This chart is a variation of the histogram , Use smooth curves to draw numerical levels , So we get a smoother distribution , And they are not affected by the number of packets used , So we can better define the shape of the distribution .

Ten 、 Population pyramid

Population pyramid (Population Pyramid) Also known as 「 Age gender pyramid 」, It's a back-to-back pair of histograms , Show the distribution of all age groups and the male and female population .

The population pyramid is best used to detect changes or differences in population patterns . Multiple population pyramids put together can be used to compare population patterns between countries or different groups .

11、 ... and 、 Bar chart

Bar chart (Bar Chart) Also known as 「 Bar chart 」 or 「 Bar charts 」, Use horizontal or vertical bars （ Bar charts ） To compare different classes of discrete values .

One of the axes in the chart represents the specific category to be compared , The other is used as a scale for discrete values .

The discrete data of bar chart is classified data , It's about the number in a single category , It doesn't show the continuous development of the value in a certain period of time .

Twelve 、 Multiple sets of bar charts

Groups of bars are also known as 「 Grouped bars 」 or 「 Double bar chart 」, It's a variation of the bar chart .

Multiple sets of bars are often used to compare grouped variables or categories with other data groups , Mini histograms can also be used to compare , Each bar in each group will represent the significant interval between variables .

But the disadvantage is , It's hard to read when there are too many bars together .

13、 ... and 、 Stacked bar chart

Unlike multiple bar charts , Stacked bar chart (Stacked Bar Graph) Overlay the bars of multiple data sets with each other , Good for showing how large categories are broken down into smaller categories , And what's the relationship between each part and the total amount .

There are two types of stacked bar charts ：

Simple stacked bar chart . Place segment values one by one , The total value of the bar is the sum of all the segment values , Suitable for comparing each group / The total number of segments .

100% Stacked bar chart . The percentage of each group as a whole will be displayed , And draw according to the percentage of each value in the whole group , It can be used to show the relative difference between the numbers in each group .

fourteen 、 Unequal width histogram

Unequal width histogram (Marimekko Chart) Also known as 「 Mosaic 」, Used to show the relationship between a pair of variables in classified data , The principle is similar to bidirectional 100% Stacked bar chart , But all the bars in it are numerical / The scale axis has equal length , And it's divided into segments .

The main disadvantage of unequal width histogram is that it is difficult to read , Especially when there are lots of segments . Besides , It's also difficult to accurately compare each segment , Because they're not aligned along a common baseline .

therefore , The unequal width histogram is more suitable for providing an overview of the data .

15、 ... and 、 Area map

Area map (Area Graph) It's a kind of line chart , But the area below the line is filled with color or texture

Just like a line chart , The area chart can show the change and development of quantitative values in a certain period of time , Most commonly used to show trends , It's not a specific number .

Two more commonly used area charts are grouped area charts and stacked area charts . The grouped area chart starts with the same zero axis , The stacked area chart starts from the last data point of the previous data series .

sixteen 、 Scale area

It's a great way to compare values and scale （ Size 、 Quantity, etc ）, In order to quickly and comprehensively understand the relative size of data , Without using a scale .

A scale area chart usually uses a square or a circle , Common technical errors are , Use length to determine shape size , Instead of calculating the area of space in a shape , This leads to an exponential increase and decrease in values .

seventeen 、 Stacked area chart

Stacked area chart (Stacked Area Graph) The principle is the same as that of a simple area graph , But it can display multiple data series at the same time , The starting point of each series is the ending point of the previous data series

Stacked area charts use area areas to represent integers , So it doesn't apply to negative values . in general , They are suitable for comparing changes in multiple variables within the same interval .

eighteen 、 Quantized waveforms

This chart is a variation of the stacked area chart , But the value is not drawn along a fixed line axis , It's around a changing central baseline

By using flowing organic shapes , Quantized waveforms (Stream Graph) Different types of data can be displayed over time , It's kind of like these organic rivers , So the quantization waveform looks pretty good .

In the quantization waveform , The shape and size of each wave is proportional to the number in each category . The axis that flows parallel to the waveform is used as a time scale . We can also use different colors to distinguish each category , Or change the color to show the additional quantitative value of each category .

Besides , When they show it interactively , More efficient than static or printed .

nineteen 、 Radar map

Radar map (Radar Chart) Also known as 「 Spiders 」、「 Polar map 」 or 「 Star map 」, It's a way to compare multiple quantitative variables , Can be used to see which variables have similar values , Or if there are any outliers in each variable .

Besides , Radar charts can also be used to see which variables in the dataset score higher / low , Ideal for display performance .

Each variable has its own axis （ Start from the center ）. All the shafts are arranged radially , The distance between them is equal , All axes have the same scale . The grid lines between the axes are usually used for guidance only . The value of each variable is drawn on its axis , All the variables in the dataset will be joined together to form a polygon .

twenty 、 Sanguitu

Sanguitu (Sankey Diagram) Used to show the flow direction and quantity .

At each stage of the process , Flow arrows or lines can be grouped together , Or separate them in different ways . We can use different colors to distinguish the different categories in the chart , Or represents a transition from one stage to another .

The 21st 、 Parallel set graph

Parallel set graphs are similar to Sankey graphs , Both show flow and scale , But parallel set graphs don't use arrows , They are in each displayed line set (line-set) Divide the process path .

Each line set corresponds to a dimension / Data sets , Its value / The different classes of line segments represented by the set of lines . The width of each line and the flow path , It is determined by the proportion of the total number of categories . Each process path can be represented by different colors , To show and compare the distribution between different categories .

Twenty-two 、 Error line

Error bars can be used as an enhancement to show data changes , It is usually used to display the standard deviation in a range dataset 、 Standard error 、 Confidence interval or minimum / Maximum .

The error line is always parallel to the axis of the quantitative scale , It can be vertical or horizontal （ Depending on the quantitative scale is in Y Axis or X On the shaft ）.

23 、 Tree structure

Tree structure (Treemap) It's a way to use nested rectangles to show hierarchies , At the same time, the number of each category is displayed by area size .

Each category is assigned a rectangular area , Its subcategories are represented by nested rectangles . When different quantities are assigned to each category , The area of these rectangles is proportional to this number .

Ben Shneiderman The tree structure diagram was originally developed to display a large number of file directories on the computer , It doesn't take up too much screen space , Therefore, the tree structure diagram is a compact and space saving hierarchical display , It allows people to quickly understand the structure .

Twenty-four 、 Circle packing

Circle accumulation (Circle Packing) Also known as 「 Circular tree structure 」, It's a variation of the tree structure , Use circles （ Instead of rectangles ） Layer by layer represents the entire hierarchy .

The area of each circle can also be used to represent any additional value , Such as quantity or file size . We can also use color to classify the data , Or represent another variable by different hues .

Although the round pile looks beautiful , But it's not as space saving as a tree （ Because there's a lot of space in the circle ）, But it's actually more effective than a tree diagram in showing hierarchies .

twenty-five 、 The pie chart

Pie chart (Pie Chart) Divide a circle into segments of different proportions , To show the proportions between the categories

The pie chart is suitable for quickly displaying the proportional distribution of data , But the main drawback is ： Can't show too many items 、 You usually need a legend 、 Can't exactly compare .

hexacosa- 、 Doughnut

Doughnut (Donut Chart) It's basically a pie chart , It's just that the middle part is cut off

however , The doughnut chart still has a slight advantage over the pie chart , It makes people no longer just look at 「 cake 」 The area of , On the other hand, we pay more attention to the change of population value ： Focus on reading the length of the arc , Instead of comparing 「 Cakes and cakes 」 There are different proportions between .

in addition , The space in the middle of the circle can be used to display other information , So it saves more space .

twenty-seven 、 Nightingale rose

Nightingale rose (Nightingale Rose Charts) Also known as 「 Polar area 」

In the Nightingale rose , What represents the numerical value is the sectional area , Not its radius .

Twenty-eight 、 Sunrise chart

Also known as 「 Multi layer pie chart 」 or 「 Radial trees 」, Show hierarchy through a series of circles , Then cut according to different types of nodes .

Twenty-nine 、 Spiral chart

Also known as 「 Time series spiral 」, Along the Archimedean spiral (Archimedean spiral) Time based data .

The chart starts at the center of the spiral and goes out . The spiral is very variable , You can use bars 、 Lines or data points , Display along the spiral path .

Spirals are great for displaying large datasets , It usually shows data trends over a long period of time , Therefore, it can effectively display periodic patterns