From the course: Complete Guide to Generative AI for Data Analysis and Data Science

Visualizing data

- [Instructor] Data visualizations are an important part of data analysis and data science, and they're tools that really fundamentally help us understand data and help us understand data quickly. They're typically used for things like showing the relationship between variables, getting an idea of what a distribution is for a particular variable, and spotting trends and other patterns in data. Now, data visualizations are important because they enable a rapid assessment of properties of the datasets and variables within those datasets, and they're also really useful for helping us communicate insights as well. Now, there are many types of visualizations. Some of the most common are used for trend analysis, and those visualizations are things like line and area charts. When we're dealing with a question of composition, pie charts and stacked bar charts are good options. If we're trying to understand the distribution of the data of a particular variable, then histograms are great, box plots, Q-Q plots, Normality, of course, is a big consideration in many statistics tests. They work well when the data's normally distributed. Sometimes we want to take a quick look at those. Histograms and Q-Q plots are good for those. Correlation, when we're trying to understand if there is a relationship between variables, such as either a positive relationship, like both variables increase together, or a negative relationship, where one increases and another decreases. So scatterplots and bubble plots can help with that. And of course, when you're looking at, like, geographic data, just mapping data on maps helps. And another type of visualization that's often used is heat maps as well. So these are different types of visualizations, and going forward, we're going to take a look at some of these different types of visualizations.

Contents