Step 1: Understand the question.
We must find the database type best suited for large amounts of unstructured data.
Step 2: Meaning of unstructured data.
Unstructured data has no fixed table shape. Examples are images, videos, emails, and social media posts. They do not fit neatly into rows and columns.
Step 3: Why tables struggle.
Relational, network, and hierarchical databases all expect a planned structure. Pouring messy, varied data into them is hard and slow.
Step 4: Meet the non-relational database.
A non-relational database, also called NoSQL, is built to hold flexible and varied data. It can store huge volumes that do not follow a fixed schema.
Step 5: Match to the need.
Since the data is large and unstructured, the flexible non-relational type is the right fit.
Step 6: Conclude.
So the answer is option 3, non-relational database.
\[ \boxed{\text{Non-relational database}} \]