In the field of data science, data visualization is undoubtedly the top word today. No matter what data you want to analyze, doing data visualization seems to be a necessary step. But many people don’t have a specific concept of data visualization, and they don’t know how to implement it. So, today I am going to take you through the definition, concept, implementation process and tools for data visualization.

1. What is data visualization?

Scientific visualization, information visualization, and visual analytics are often seen as the three main branches of visualization. The new discipline “Data Visualization”, which is a combination of these three branches, is a new starting point in the field of visual research.
Generalized data visualization involves various disciplines such as information technology, natural science, statistical analysis, graphics, interaction, and geographic information.

1.1 Scientific Visualization

Scientific visualization is an interdisciplinary research and application field in science, focusing on the visualization of three-dimensional phenomena, such as architecture, meteorology, medicine or biological systems. Its purpose is to graphically illustrate scientific data, enabling scientists to understand, explain, and collect patterns from the data.

1.2 Information visualization

Information visualization is the study of interactive visual representations of abstract data to enhance human cognition. Abstract data includes both digital and non-digital data such as geographic information and text. Graphics such as histograms, trend graphs, flow charts, and tree diagrams all belong to information visualization, and the design of these graphics transforms abstract concepts into visual information.

1.3 Visual Analytics

Visual analytics is a new field that has evolved with the development of scientific visualization and information visualization, with an emphasis on analytical reasoning through an interactive visual interface.
From FineReport

2. Why do we need data visualization?

The amount of information that humans gain through vision is far beyond that of other organs. Data visualization is the use of human natural skills to enhance data processing and organization efficiency.
Visualization can help us deal with more complex information and enhance memory. Most people don’t know much about statistical data, and basic statistical methods (mean, median, range, etc.) are not in line with human cognitive nature. One of the most famous examples is Anscombe’s quartet. It is difficult to see the law according to statistical methods, but the rules are very clear when the data is visualized.

3. How to achieve data visualization?

Technically, the simplest understanding of data visualization is the mapping from data space to graphic space.
A classic visual implementation procedure is to process and filter the data, transform it into an expressible visual form, and then render it into a user-visible view.

Visualization technology stack

In general, professional data visualization engineers need to master the following technology stack: · Basic mathematics: trigonometric function, linear algebra, geometric algorithm · Graphics: Canvas, SVG, WebGL, computational graphics, graph theory · Engineering algorithms: basic algorithms, statistical algorithms, common layout algorithms · Data analysis: data cleaning, statistics, data modeling · Design aesthetics: design principles, aesthetic judgment, color, interaction, cognition · Visual basis: visual coding, visual analysis, graphical interaction · Visualization solutions: correct use of charts, visualization of common business scenarios

4. Common data visualization tools

Generally speaking, R language, ggplot2 and Python are used in academia. The most familiar tool for ordinary users is Excel. Commercial products include Tableau, FineReport, Power BI, etc.

1) D3

D3.js is a JavaScript library based on data manipulation documentation. D3 combines powerful visualization components with data-driven DOM manipulation methods.
Evaluation: D3 has powerful SVG operation capability. It can easily map data to SVG attribute, and it integrates a large number of tools and methods for data processing, layout algorithms and calculating graphics. It has a strong community and rich demos. However, its API is too low-level. There isn’t much reusability while the cost of learning and use is high.

2) HighCharts

HighCharts is a chart library written in pure JavaScript that makes it easy and convenient for users to add interactive charts to web applications. This is the most widely used chart tool on the Web, and business use requires the purchase of a commercial license.
Evaluation: The use threshold is very low. HighCharts has good compatibility, and it is mature and widely used. However, the style is old, and it is difficult to expand charts. And the commercial use requires the purchase of copyright.

3) Echarts

Echarts is an enterprise-level chart tool from the data visualization team of Baidu. It is a pure Javascript chart library that runs smoothly on PCs and mobile devices, and it is compatible with most current browsers.
Evaluation: Echarts has rich chart types, covering the regular statistical charts. But it is not as flexible as Vega and other chart libraries based on graphic grammar, and it is difficult for users to customize some complex relational charts.

4) Leaflet

Leaflet is a JavaScript library of interactive maps for mobile devices. It has all the mapping features that most developers need.
Evaluation: It can be specifically targeted for map applications, and it has good compatibility with mobile. The API supports plug-in mechanism, but the function is relatively simple. Users need to have secondary development capabilities.

5) Vega

Vega is a set of interactive graphical grammars that define the mapping rules from data to graphic, common interaction grammars, and common graphical elements. Users can freely combine Vega grammars to build a variety of charts.
Evaluation: Based entirely on JSON grammar, Vega provides mapping rules from data to graphics, and it supports common interaction grammars. But the grammar design is complex, and the cost of use and learning is high.

6) deck.gl

deck.gl is a visual class library based on WebGL for big data analytics. It is developed by the visualization team of Uber.
Evaluation: deck.gl focuses on 3D map visualization. There are many built-in geographic information visualization common scenes. It supports visualization of large-scale data. But the users need to have knowledge of WebGL and the layer expansion is more complicated.

7) Power BI

Power BI is a set of business analysis tools that provide insights in the organization. It can connect hundreds of data sources, simplify data preparation and provide instant analysis. Organizations can view reports generated by Power BI on web and mobile devices.
Evaluation: Power BI is similar to Excel’s desktop BI tool, while the function is more powerful than Excel. It supports for multiple data sources. The price is not high. But it can only be used as a separate BI tool, and there is no way to integrate it with existing systems.

8) Tableau

Tableau is a business intelligence tool for visually analyzing data. Users can create and distribute interactive and shareable dashboards, depicting trends, changes and densities of data in graphs and charts. Tableau can connect to files, relational data sources and big data sources to get and process data.
Evaluation: Tableau is the simplest business intelligence tool in the desktop system. It doesn’t force users to write custom code. The software allows data mixing and real-time collaboration. But it’s expensive and it performs less well in customization and after-sales services.

9) FineReport

FineReport is an enterprise-level web reporting tool written in pure Java, combining data visualization and data entry. It is designed based on “no-code development” concept. With FineReport, users can make complex reports and cool dashboards and build a decision-making platform with simple drag-and-drop operations.
Evaluation: FineReport can be directly connected to all kinds of databases, and it is convenient and quick to customize various complex reports and cool dashboards. The interface is similar to that of Excel. It provides 19 categories and over 50 styles of self-developed HTML5 charts, with cool 3D and dynamic effects. The most important thing is that its personal version is completely free.

Conclusion

Data visualization is a huge field with many disciplines. It is precisely because of this interdisciplinary nature that the visualization field is full of vitality and opportunities.
Originally published at https://medium.com/@lewischou62 
Follow FineReport Reporting Software on Facebook to learn more about data visualization tools!