Shulou(Shulou.com)06/01 Report--

What is the basic process of data visualization? I believe many inexperienced people are at a loss about it. Therefore, this paper summarizes the causes and solutions of the problem. Through this article, I hope you can solve this problem.

What we want is not the data, but the facts the data tells us. Most people face the challenge that we recognize the need for data visualization, but lack expertise in data visualization. Part of the reason can be attributed to the fact that data visualization is only one link in the process of data analysis, and data analysts may spend their energy on obtaining data, cleaning and collating data, analyzing data, and building models. But it is not up to the ultimate display and communication.

This is part of the reason why "code writers can't do PPT". In fact, as long as we master the visualization skills, our work is more likely to be recognized by leader.

Visualization tools include, but are not limited to, Tableau,Excel,PowerBI,Python,R

Before Visualization: exploratory Analysis and explanatory Analysis

There is an important difference between the two: exploratory analysis means understanding the data and finding the essence that is worth analyzing or sharing with others. It's like looking for pearls in an oyster, you may open a hundred oysters (try many ways) to finally find two pearls. And interpretive analysis, we are eager to be able to say something, tell a good story-focus on two pearls.

Most of the time, our reporting work is to do a good job of interpretive analysis.

Visualization process

A complete data visualization process, which mainly includes the following four steps:

Determine the theme of data visualization

Extract data from visual themes

Determine the chart according to the data relationship

Visual layout and design

Picture from: Mudong Koji

The visual element consists of three parts: visual space + tag + visual channel.

Visualization space

The display space of data visualization, usually two-dimensional. The visualization of three-dimensional objects, through graphics rendering technology, solves the problem of display in two-dimensional plane, such as 3D ring graph, 3D map and so on.

Marking

Tag, which is the mapping of data attributes to visual geometry elements, is used to represent the classification of data attributes.

According to the difference of spatial degrees of freedom, markers can be divided into point, line, surface and volume, which have zero degree of freedom, one-dimensional, two-dimensional and three-dimensional degrees of freedom respectively. For example, our common scatter chart, line chart, rectangular tree chart and three-dimensional column chart use four different types of marks: point, line, surface and volume.

Visual channel

The mapping of the value of the data attribute to the visual rendering parameter of the tag, called the visual channel, is usually used to show the quantitative information of the data attribute.

Common visual channels include: location, size (length, area, volume), shape (triangle, circle, cube), direction, color (hue, saturation, brightness, transparency, etc.).

Determine the chart

The interrelationship between the data determines the types of charts that can be used. The correspondence between common data relationships and chart types is shown in the following figure:

In future columns, we will gradually share the graphic application cases and precautions that appear in the above pictures. Next, we combine with specific cases to illustrate the charm of data visualization.

Form

One thing to keep in mind when using tables is to integrate the design into the background and keep the data at the core. Don't let the heavy borders and shadows compete with the data for the audience's attention. Instead, use spaces to distinguish the elements in the table.

The left table with too many frame lines can easily disturb the reader's attention. on the contrary, the three-line table is concise and capable and is usually the first choice for the table style of papers and publications.

Thermal map

Thermal map is a method of visualizing data in the form of a table, where the data is displayed (outside the data) using colored cells to transmit information about the relative size of the data.

Thermal map drawing (excel2010 version): select the data-start-style-conditional format (you can choose the appropriate conditional format as needed).

Line chart

We can usually use discount maps to understand trends, for example, the annual rainfall of the time series (the sum of the daily rainfall); in some cases, the lines in the line chart may represent a comprehensive statistic, such as an average or a predicted point estimate. If you still want to show the range (or confidence interval), you can visualize it directly on the diagram.

The left is a multi-indicator line chart, and the right shows the average in the range in the line chart.

When drawing the right picture: draw the avg--max line chart first, then right-click "change Chart Type" and select "area Chart"

Right-click "Select data", add min line chart; finally, select "min discount chart", right-click "set data series format", and select "solid color fill-white".

After reading the above, have you mastered the basic flow of data visualization? If you want to learn more skills or want to know more about it, you are welcome to follow the industry information channel, thank you for reading!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.