Visualizing data is a step that gets overlooked by data scientists. It helps us tell stories by analyzing and curating data into a form easy to understand. By removing all the technical detail and noise and highlighting key information, data scientists can explain the importance of their work to non-technical managers and executives.
There are many tools to help visualize data. For years, Microsoft Excel dominated the static visualization market. Over time, we gravitated to dynamic visualizations and flexibility to showcase more data in a cleaner manner. Two types of tools helped create dynamic visuals.
- Business Intelligence and Analytics Software: Tableau, PowerBI
- Open-sourced programming libraries: D3.js, Plotly Dash
Third party software tools like Tableau and PowerBI are excellent for non-technical folks. Drag and drop interfaces and abstractions allow analysts to create dynamic dashboards easily. The drawbacks are
- software tools are expensive
- a bit of a learning curve to learn these tools
- limits to visualization design; software may not allow some components
Open-sourced programming libraries are excellent for technical folks. Those comfortable with software engineering can follow the documentation to create flexible dynamic visualizations with ease. Furthermore, these packages are free to use (with Plotly offering a paid version for its enterprise Dash components).
The difference between D3.js and Plotly are the following
- D3.js is designed in JavaScript, Plotly is designed in Python
- D3.js has been around longer than Plotly, and thus has better community support and a more mature ecosystem(extensive examples and tutorials).
- D3.js requires engineers to understand the low-level details of web development (HTML, CSS, JavaScript) in order to use it effectively. Plotly abstracts such details in simple-to-use Python classes.
- D3.js has a steep learning curve due to its JavaScript nature…
Be the first to comment