radar diagram is a graphical method for displaying multivariate data in the form of a two-dimensional chart of three or more quantitative variables represented on the axis starting from the same point. The relative position and the angle of the axis are usually not informative.
Radar graphs are also known as web graphs , spider charts , star charts , star plot , cobweb diagrams , irregular polygons , polar diagrams , or Kiviat charts . This is equivalent to a parallel coordinate plot in polar coordinates.
Video Radar chart
Ikhtisar
The radar graph is a graph and/or plot consisting of a series of equi-angular radii, called radii, with each speaking representing one of the variables. The data length of a speech is proportional to the magnitude of the variable for the data point relative to the maximum magnitude of the variable across all data points. The drawn line connects the data values ​​for each speech. It gives the appearance plot like a star and the origin of one of the popular names for this plot. The star plot can be used to answer the following questions:
- Which observations are most similar, ie, is there a collection of observations? (The radar graph is used to check the relative value for a single data point (eg, 3 large points for variables 2 and 4, small for variables 1, 3, 5, and 6) and to find the same point or different point. li>
- Is there anything outside?
Radar charts are a useful way to display multivariate observations with a number of random variables. Each star represents one observation. Typically, radar graphs are generated in multi-plot format with many stars on each page and each star represents one observation. The star plot was first used by Georg von Mayr in 1877. The radar chart differs from the glyph plot in which all the variables are used to construct a plotted star figure. There is no separation to the foreground and background variables. Instead, the stars are usually arranged in a rectangular arrangement on the page. It is somewhat easier to see patterns in data if observations are organized in some non-arbitrary settings (if the variables are assigned to starlight in some meaningful sequence).
Maps Radar chart
Apps
One application of a radar diagram is an improved quality control to show performance metrics of each program in progress.
They are also used in sports to map the strengths and weaknesses of players, where they are usually called spider charts.
Limitations
The radar graph is particularly suitable for displaying nine and similarities , or when one graph is larger in each variable than the other, and is mainly used for ordinal measurements - where each variable corresponds to "better" in some ways, and all variables on the same scale.
In contrast, radar graphs have been criticized for being less suited to making trade-off decisions - when one graph is larger than the other on some variables, but less on the other.
Furthermore, it is difficult to visually compare the lengths of different radii, since radial distances are difficult to assess, although concentric circles help as grid lines. Instead, one can use simple line graphs, especially for time series.
Artificial structure
Radar charts impose some structure on the data, which is often contrived:
- Neighbor linkage - radar diagrams are often used when neighboring variables are unrelated, creating fake connections.
- The cyclic structure - the first and last variables are placed next to each other.
- Long - often the most natural variables ordinal: for better or worse, though degree difference may be artificial.
- Area - area scale as squared value, exaggerates the effect of large numbers. For example, 2, 2 requires 4 times the area of ​​1, 1. This is a common problem with the graphics area, and the area is difficult to assess - see the "hierarchy of Cleveland".
For example, alternating data 9, 1, 9, 1, 9, 1 produce a spiking (incoming and outgoing) radar graph, while rearranging the data as 9, 9, 9, 1, 1, 1 instead produces two distinct slices ).
In some cases there is a natural structure, and radar graphics can be very appropriate. For example, for data diagrams that vary over a 24-hour cycle, the hourly data is naturally related to its neighbors, and has a cyclic structure, so it can naturally be displayed as a radar diagram.
A set of guidelines on the use of radar charts (or rather "polar area graphs" closely related) are:
- You do not mind reading the stacked area instead of the position along the scale (see Cleveland Hierarchy),
- The data set is completely cyclic, is not linear, and
- there is a two series to compare, which is much smaller than others
Data set size
Radar graphs are helpful for small to medium sized multivariate data sets. Their main drawback is their effectiveness is limited to data sets with fewer than a few hundred points. After that, they tend to be extraordinary.
Example
The chart on the right contains a star of 15 cars. The variable list for an example star plot is:
- Price
- Distance Distance (MPG)
- 1978 Record of Repair (1 = Worst, 5 = Best)
- Record of Improvement 1977 (1 = Worst, 5 = Best)
- Headroom
- Backseat Space
- Space
- Weight
- Long
We can view these plots separately or we can use them to identify groups of cars with similar features. For example, we can see the plot of Cadillac Seville star (the last one on the picture) and see that it is one of the most expensive cars, gets below average (but not among the worst) gas mileage, has an average fix record, and average room and average size above average. We can then compare the Cadillac model (the last three plots) with the AMC model (the first three plots). This comparison shows a different pattern. The AMC model tends to be inexpensive, has gas mileage below average, and is small in both height and weight and indoors. Cadillac models are expensive, have poor mileage, and are large and wide.
Alternative
Simply put, one can use simple line graphs, especially for time series.
For a qualitative comparison of graphical 2-dimensional tabular data in several variables, a common alternative is Harvey Balls, which is used extensively by Consumer Reports. Comparisons in Harvey Balls (and radar graphs) can be significantly assisted by ordering variables algorithmically to add orders.
The best way to visualize structures in multivariate data is offered by major component analysis (PCA).
Another alternative is to use a bar graph, which can be compared with a mini graph.
Although radar and polar graphs are often depicted as the same type of graph, some sources make a difference between them and even consider a radar graph as a variation of a polar graph that does not display data in terms of polar coordinates.
See also
- Plot (graph)
- Polar area diagram
- Parallel coordinates
References
This article incorporates public domain material from the National Institute of Standards and Technology website http://www.nist.gov.
External links
- Star Plot - NIST/SEMATECH e-Handbook from Statistical Method
Source of the article : Wikipedia
0 Comments