- a vector is a list of numbers that represents something.
- used to encode data points in machine learning so that algorithms can process more efficiently.
- extremely important in [[{3.1} the algorithm is a giant system|recommendation systems]].
- vectors can be used to represent:
- **text** → words or documents can be represented as vectors, e.g. where each number represents the importance of a term in a file, or some other semantic meaning.
- **images** → components of the vector represent each pixel; a 28x28 grayscale image would have 784 numbers in its vector.
- **features** → where each attribute is a component of the vector. for example:
| house id | bedrooms | bathrooms | sq ft | vector representation |
| -------- | -------- | --------- | ----- | --------------------- |
| 001 | 3 | 1 | 1500 | [3, 1, 1500] |
| 002 | 2 | 2 | 1500 | [2, 2, 1500] |
| 003 | 3 | 2 | 1750 | [3, 2, 1750] |