Every day, we produce enormous amounts of data. In actuality, the last three to four years have seen a 90% increase in global data production! The figures are simply astounding. Here are just a few instances of the types of information being gathered:

- Facebook gathers information on your likes, shares, posts, whereabouts, favourite restaurants, and other activities.
- Many pieces of personal data about you are gathered by your smartphone apps.
- Amazon gathers information about your site activity, including what you browse, click, and buy.
- Each customer's every step is monitored by casinos.

It gets harder and harder to visualise data and make judgments as its generation and gathering grow. Charts are one of the most used tools for visualisation.

The following are some advantages of using dimensionality reduction on a dataset:

- The less dimensions there are, the less space is needed to hold the data.
- Less dimensions result in faster computation and training
- Some algorithms struggle to function properly when the dimensions are huge. Therefore, it is necessary to reduce these dimensions for the algorithm to be effective.
- By deleting extraneous characteristics, multicollinearity is handled. For instance, you have two variables: "calories burned" and "time spent on treadmill in minutes." These two factors are closely related since the longer you run on a treadmill, the more calories you'll burn. Therefore, there is no use in storing both as you can get by with only one of them.