Basically it is based on statistical and time-series analysis of data based on that you may be able to predict something. Just create some data-set and try to find your result at IBM Watson analytics.
Big data has meaning ful data hidden inside it. These analytics are to be good enough to extract meaningful or useful data (which will be smaller in size) to support business intelligence. People are now migrating to thick data now.
Everyone today is connected to internet and electronic devices. Our activities with them generate a lot of data every second. Like what we shop, our shopping behaviours, the places we visit, the pattern of products we use etc. This information is generated when we use internet or visit places like hospitals, shppoing malls etc.
Data is also generated from the non human activities like business trends, airplain, electric pole etc.
Read here what are the applications of this data - https://data-flair.training/blogs/big-data-applications-various-domains/