Machine Learning in Java - Second Edition

Finding or observing data

Data can be found or observed in many places. An obvious data source is the internet. With an increase in social media usage, and with mobile phones penetrating deeper as mobile data plans become cheaper or even offer unlimited data, there has been an exponential rise in data consumed by users.

Now, online streaming platforms have emerged—the following diagram shows that the hours spent on consuming video data is also growing rapidly:

To get data from the internet, there are multiple options, as shown in the following list:

• Bulk downloads from websites such as Wikipedia, IMDb, and the Million Song Dataset (which ...