DATA VISUALIZATION 101
Data visualization is growing in popularity, but there is a wide range of opinions on the requirements for creating one. This talk and class will focus on data visualization from the perspective of a data scientist and present real world examples from both client engagement and research findings. In this class, you will learn to:
- Understand basic principles essential to creating high quality and reliable data visualizations
- Understand the skills necessary to both create and evaluate data visualizations Have a basic understanding of key concepts essential to gaining effective insights from data analysis and visualization
- Strengthen the audience's ability to assess, evaluate and create accurate visualizations
- Learn a list of pitfalls to look for when working with data visualizations and what can be done to avoid being lead astray by data
- The ultimate goal is to inspire a new, more rigorous approach to applying visualization methods
WHAT IS DATA VISUALIZATION?
Data science is a new and growing field that offers an exciting opportunity for anyone who is interested in combining technical capabilities with an interest to test and apply theoretical models to solve interesting real world problems.
The focus of the talk will be on introducing the conceptual framework for creating optimal data visualizations and a technical background is not required to benefit from attending the presentation.
The talk will begin by discussing the present state of data visualization and present an overview of the most important concepts and key considerations to keep in mind when creating data visualizations. Next, we will introduce different approaches for creating, using, and evaluating data visualizations for a wide range of purposes.
The audience will also learn common pitfalls seen in data visualization. Real world examples will be presented to demonstrate each pitfall, show why they occurred with strategies to avoid them and will serve to strengthen the audience’s understanding of how data visualizations fit into the world of data science.
One of the most interesting examples will be from my recent investigation into the safety of a particular medication using the FDA’s Adverse Events Reporting System (AERS), which contains over 4.2+ million individual case reports from 2004 to 2012. The research not only introduces new work in the field of data science, but the findings that are presented highlights potential issues regarding the safety of drugs that are currently on the market.