Banner Default Image

The What, Why and How of Data Visualisation

22 days ago by Jonathon Webley

The What, Why and How of Data Visualisation

I Stock 547419538 Small

Data Science is the buzz word of the moment and the topic that we see discussed a lot - but data science is nothing without data visualisation. It doesn’t matter what type of data you are wanting to analyse, or how much; you need data visualisation in order to do this successfully. However, not everyone in the world of business has knowledge of this area of big data, and so in this article, we are going to look at the what, why and how of data visualisation.

What is data visualisation?

In the past, there have been three main branches of visualisation identified: information visualisation, scientific visualisation and visual analytics.

Information visualisation is the study of visual representations of abstract data which are usually interactive. Abstract data can include both digital and non-digital data such as text and geographic information. This data is represented graphically through histograms,
glow charts, trend graphs and tree diagrams, to help transform abstract concepts into visual information.

Scientific visualisation is a field of science which is concerned with the visualisation of 3D phenomena such as biological systems, medicine and meteorology. It is used to illustrate scientific data in a graphical way so that scientists can understand and collect patterns from the data - as well as explain it more easily.

 Visual data is a fairly new field of visualisation that has evolved from information and scientific visualisation, and which has more emphasis placed on analytical reasoning through the use of an interactive visual interface.

With the rise in the use of data recently, there is now a new branch called data visualisation - which is a combination of the other three branches and is used as a starting point in the field of visual research.

In general data visualisation covers various disciplines including:   

  • Graphics   

  • Geographic information  

  • Information technology

  • Interaction 

  • Natural science  

  • Statistical analysis

Why do we need data visualisation?

As humans, we gather a huge amount of information through the use of our eyes - far more information that we gather through any other organ in our body. Data visualisation allows us to use our natural skills of sight to enhance our data processing skills by dealing
with more complex information and therefore improves our organisational efficiency.

A good example of this is Anscombe’s quartet which is a well-known statistical data method comprising of four data sets with nearly identical simple descriptive statistics, but which look very different when they are placed into graph form.

How is data visualisation achieved?

The simplest way to achieve data visualisation is to map data to the graphic space. So firstly, you need to process and filter the data, then transform it into a visual form, and then render it into a user-visible view.

If you are working as a data visualisation engineer, you would need to be proficient in the following:

  • Basic maths skills such as linear algebra, geometric algorithms, trigonometric functions

  • Data aesthetics such as aesthetic judgment, cognition, colour, design principles, interaction

  • Data analysis such as data cleaning, data modelling, and statistics

  • Engineering algorithms such as basic algorithms, common layout algorithms and statistical algorithms

  • Visual Basis - visual analysis, visual coding, graphical interaction

  • Visualisation solutions - correct use of charts and visualisation of common business scenarios

 What are the most common data visualisation tools used?

The most common data visualisation tools will depend on what area of industry you are working in. For example, most ordinary users will opt for Excel, whereas academic users may opt for R language or Python, and commercial users tend to use Power BI, Tableau or Fine Report.

Having said that, some of the most popular tools used include:

  • D3 - a JavaScript library which combines data-driven DOM manipulation methods with powerful visualisation components

  • deck.gl - a visual class library based on WebGL for big data analytics

  • Echarts - a pure JavaScript chart library which runs smoothly on PCs and mobile devices and is compatible with most current browsers      

  • FineReport - an enterprise-level web reporting tool written in pure Java and designed based on a no-code development concept

  • HighCharts - the most widely used chart tool on the web, which is written in JavaScript, making it easy for users to add interactive charts to their web applications

  • Leaflet - a JavaScript library of interactive maps for mobile devices

  • Power BI - similar to Excel’s desktop Bi tool but supports multiple data sources, simplifying their preparation and providing instant analysis

  • Tableau - simplest business intelligence tool which doesn’t force users to write custom code, but which can connect to files, big data sources and relational data sources to create and distribute interactive and shareable dashboards

  • Vega - Based on JSON grammar, providing mapping rules from data to graphics and supporting common interaction grammars.

As you can see, data visualisation is a huge field with many disciplines - which makes it packed full of opportunities.

If you are looking for permanent or contract hires in the field of data visualisation, or you want to talk to us about your career in data visualisation then email us on info@agilerecruit.com or call us on 0161 416 6633.