Data Visualization
Data visualization refers to the use of graphical means to display data in order to conduct more intuitive and in-depth observation and analysis of the data. It is a technical study of the visual representation of data. “A picture is worth a thousand words” emphasizes the importance of visualization in expressing information.
1. Classification and composition of figures for scientific research papers
| tool | Features |
|---|---|
| Python | Provides rich data analysis and visualization libraries, such as Matplotlib, Seaborn and Plotly. Supports drawing line charts, scatter charts, column charts, etc. High degree of customization. |
| R | Programming language for statistical computing and graphics. With powerful data processing and visualization capabilities, the ggplot2 package provides a rich graphics library. Supports drawing scatter plots, box plots, area charts, etc. |
| MATLAB | Professional numerical calculation and visualization software, suitable for scientific engineering fields. Provides powerful data processing and drawing functions. Supports drawing curve graphs, polar coordinate graphs, contour graphs, etc. |
| Excel | Common office software provides simple data analysis and drawing functions. Suitable for simple data visualization, such as line charts, bar charts, scatter charts, etc. |
| SPSS | Statistical analysis software for data processing, statistical modeling and visualization. Provides graphics options and customization capabilities. Supports drawing bar charts, pie charts, radar charts, etc. |
| Origin | Professional scientific drawing software widely used for data analysis and visualization. Provides a variety of graph types and drawing tools, supporting drawing curve charts, area charts, 3D charts, etc. |
Format and size of figures for scientific research papers
| Font settings | describe |
|---|---|
| Serif | Serif fonts, with distinct stroke weight differences, are suitable for body text. Common serif fonts include Times New Roman and Georgia. |
| Sans serif | Sans serif font with consistent stroke thickness, suitable for titles, chart labels, legends, etc. Common sans serif fonts include Arial and Helvetica. |
| Minivan | Monowidth font, each character occupies the same width, suitable for displaying codes, commands and content arranged in a fixed width. Common fixed-width fonts include Courier New and Consolas. |
| Font size setting | describe |
|---|---|
| small | Usually used for legends, chart labels, diagrams and other smaller text information. You can use font size of 10pt or 12pt. |
| normal | Suitable for body text and general chart text. You can use font size of 12pt or 14pt. |
| large | Used to highlight titles and important information. The font size you can use is 14pt or 16pt. |
In scientific research papers, in order to maintain visual unity and readability, it is recommended to use the same font and size settings throughout the paper. For titles, paragraphs, diagram labels, legends, etc., you can select appropriate fonts and font sizes to ensure text clarity and overall aesthetics.
Drawing principles
| in principle | describe |
|---|---|
| principle of necessity | Determine whether it is necessary to include figures in scientific research papers |
| Conduct specific problem analysis | |
| Pictures are necessary when they serve to supplement explanatory text, visually display results, and lead to the following. | |
| Avoid too many illustrations and improve your ability to select illustrations | |
| legibility principle | Comply with the principles of legibility when drawing illustrations |
| Complete, accurate titles, labels, and legends enhance legibility | |
| consistency principle | The picture content is consistent with the context or specified content description |
| The abbreviations and symbols of physical quantities should be consistent with the text of the paper. | |
| Illustration data is consistent with the context | |
| Illustration scale and scaling remain consistent | |
| The elements of each layer of similar images should be consistent |
color mode
| color mode | describe |
|---|---|
| RGB | Express various colors by mixing three colors: red, green, and blue |
| Use the changes in the three color channels of red, green, and blue and their superposition to obtain various color values | |
| Assign red, green, and blue intensity values to each pixel in the image (the value range is 0 to 255) | |
| Black is represented as (0,0,0), white is represented as (255,255,255) | |
| Red, green, and blue are located at the vertices of the three-dimensional cube, black is at the origin, white is at the vertex farthest from the origin, and yellow, magenta, and cyan are located at the remaining three vertices. | |
| The closer a different color is to a black vertex, the darker the color; the closer it is to a white vertex, the lighter the color. | |
| CMYK | Use 4 ink colors of Cyan, Magenta, Yellow, and Black to express various colors |
| Various colors can be obtained by superimposing cyan, magenta, yellow and black (transparent ink) | |
| Commonly used in the printing industry, suitable for four-color printing processes | |
| HEX | How to express colors using hexadecimal |
| Use a 6-digit hexadecimal value to represent the color. The first two digits represent the red component, the middle two digits represent the green component, and the last two digits represent the blue component. | |
| The value range of each component is 00 (minimum value) to FF (maximum value), with a total of 256 levels | |
| Commonly used in various digital design tools and web design |
coloring scheme
| coloring scheme | describe |
|---|---|
| monochrome | Color combinations with the same or similar hue, obvious light and dark levels, and the same saturation |
| complementary colors | Choose colors whose hues are opposite each other on the color wheel for a contrasting effect |
| isometric triangle | Choose 3 colors that are 120° apart from each other on the color wheel, one as the main color and two as secondary colors |
| four corners | One is a matrix color scheme consisting of two pairs of complementary colors, and the other is a square color scheme |
color theme
| 不同的绘图工具颜色主题 | 描述 |
|---|---|
| Matplotlib 默认颜色主题 | Matplotlib库的颜色主题包括单色系（sequential）、双色渐变色系（diverging）和多色系（qualitative）等。单色系主题中颜色的色相基本相同，饱和度单调递增。双色渐变色系主题主要用在有一个关键中心值的数值变量中。多色系主题适用于表示类别型数值。 |
| Seaborn 默认颜色主题 | Seaborn库提供了丰富的颜色主题选项，用于数据可视化。Seaborn的默认颜色主题具有较高的色彩亮度和明亮的色调，适用于绘制图表、统计图和复杂的数据可视化。 |
| SciencePlots 默认颜色主题 | SciencePlots是一个科学绘图主题的集合，提供了多个专为科学论文配图设计的默认颜色主题。这些主题通常符合期刊的配色要求，用户可以直接使用它们来保证图表的美观并满足期刊的要求。 |
绘制工具
| 颜色工具 | 描述 | 特点 |
|---|---|---|
| Color Scheme Designer | Color Scheme Designer 是一个高级的在线配色器网站。 | 1. 允许用户基于不同的颜色模型（如RGB、CMYK和HSL）创建配色方案。 |
| 2. 提供单色、互补和三元配色方案选项。 | ||
| 3. 可以预览和优化配色方案。 | ||
| Adobe Color | Adobe Color 是Adobe提供的一款探索和灵感工具。 | 1. 允许用户基于类似色彩、单色和互补等不同的颜色规则创建调色方案。 |
| 2. 与Adobe Creative Cloud应用程序集成，方便访问配色方案。 | ||
| 3. 使用户能够浏览和保存社区共享的调色方案。 | ||
| ColorBrewer 2.0 | ColorBrewer 2.0 是一款专为制图和数据可视化设计的颜色选择工具。 | 1. 提供针对不同数据类型的预设计配色方案，如顺序、发散和定性配色。 |
| 2. 提供可配置的颜色数量选项。 | ||
| 3. 确保配色方案对色盲友好，并能有效传递信息。 |
资料参考：https://github.com/datawhalechina/paper-chart-tutorial