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

How to choose a data visualization tool?

How to choose data visualization tools? Before you answer this question, you now need to answer another question, what you need to do with these data visualization tools and what to achieve.

Maybe you have a complete idea, which has been verified, and needs to be presented in a more intuitive and easy-to-understand way to tell a logic or a story; maybe you have a lot of data, and how do you want to discover and mine it? and show some of the knowledge or insights behind the data. Maybe you have all kinds of data, but you don't understand data modeling, programming, or data cleaning, or even SQL optimization. You need an easy-to-use data visualization tool to visualize the data by dragging and dropping, and you can give the most appropriate display graphics. There may be a variety of other scenarios, but all data visualization tools have a scenario for their core services, beautiful, easy to use, simple, collaborative, intelligent, and so on. Each data visualization tool has a positioning label. Choices should be made based on the core needs we need. Make a simple classification:

1. Easy-to-use, diverse presentation tools with clear goals, such as Tableau

two。 Can support flexible and customized display types, such as icon library D3

3. Data explorers with unclear targets, such as google spreadsheet's explore

4. There are industry demands that can both visual analysis and data exploration, such as sailing soft FineBI

5. Data visualization based on industry or functional requirements, such as DOMO,Qlikview

Previously, I saw that a netizen used 24 tools to make the same chart, compared 12 visualization software and 12 programming / chart libraries, and wrote a first-class article on the four aspects of tool / chart library: side weight, flexibility, chart innovation and interactive effect.

In May, the girl set herself a challenge: to try to visualize data in as many programming languages or software as possible. In order to compare these tools, she used them to repeat the same scatter chart. Based on the results, she also published two articles: one is to make the same chart with 12 kinds of software, and the other is to make the same chart with 12 kinds of programming / chart library. The following figure shows her making the same scatter chart using 12 different software:

This is the effect of 12 programming / chart libraries:

She learned from these visualization software / chart libraries that there are no perfect tools, but if you set a goal, you can find the right tool to achieve it. Here are some of the contradictions she has encountered in her production, which are often encountered by data visualizers.

1. Analyze the VS display:

Do you want to use tools (R, Python) to analyze data, or do you want to focus more on building visualization (D3.js, Illustrator)? Some BI tools (such as FineBI, Tableau, Plotly) try to strike a balance between them, which can be both analyzed and demonstrated. She arranges visualization tools and programming languages according to the side of analysis and presentation: you can see that tool classes tend to pay more attention to presentation, while programming classes are more average and have their own emphasis.

2. Data management

What if you need to change the source data when making visualization? How flexible are these tools or programming languages in this regard?

Low flexibility: in Illustrator, for example, even if you make minor changes to the data, you need to start making charts all over again, and this tool is not convenient for data management.

Psychic activity: in D3.js, for example, you can process or modify the data individually, and then re-import the data file to update the visualization results.

High flexibility: for example, in FineBI, data analysis processing such as data modeling, data cleaning, and even SQL optimization, a large amount of data processing can be done in one platform, at the same time easy to use, drag and drop to complete data visualization.

3. Traditional chart VS innovative chart:

If you only need basic chart types, such as bar charts or line charts, Excel is fine, but if you want to create interactive charts with richer forms, such as clicking to show cool interactive effects, programming languages like D3.js are more suitable, but the barriers to learning such tools are often higher, with steep learning curves and lengthy code. Or you can use Processing, which can be used to make this scatter chart with code half the length of D3.js. There is also Lyra, which does not require any code base, but also allows you to easily modify visual elements related to the data.

4. Interactive chart VS static chart:

Do you need to create web-based interactive diagrams (such as D3.js, what Highcharts can do), or will PDF/SVG/PNG-based diagrams satisfy you (R and Illustrator can do that)? Interactive diagrams were highly sought after a few years ago, but now the focus is slowly shifting from "how does it look" to "what makes more sense". For the analysis part, interactive features are often necessary. Plotly and R's library Ggvis allows readers to easily hover over visual elements to view basic data. The following figure shows the author's static and interactive division of software / programming:

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.