Looking for similar data?

Question

data mining

One evening, as is often the case, an idea came to mind. But for its implementation, my knowledge of self-learning systems is clearly not enough.
Simply put, there is a large array of diverse data. The format of the data is not yet clear. These will be either histograms or arrays of key points.
The task is to take a certain element of this array and find all elements approximately similar to it.
The data is generated from images. Unfortunately, I dealt only with classification. I will be grateful for help.

Answer 1

google: nearest neighbor search

Answer 2

Hadoop in your hands