Almost that.
The concept of binary search is the idea of dividing the search space successively into halves until you only have the element you were looking for or conclude that it is not in the set of wanted elements. To work, it is assumed that this search space has the elements properly ordered so that when breaking it in the middle, you know in which remaining half to continue the search.
However, the concept of binary search says nothing about the implementation itself. It can be files, arrays, lists, playing cards in a person’s hand or even a sequence of stones sorted by size on a table.
The binary search tree is a particular implementation. The binary tree is a very versatile and efficient data structure (as long as they are at least reasonably balanced trees) to store data that allows later searching by binary search.
However, it is not that the binary tree is a modification of the binary search algorithm. It is actually a strategy to be used to attack a problem. If your problem depends on finding data quickly and these can be ordered by some criterion, one way to achieve this is by organizing them into a binary tree so you can do the binary search.
It reminded me that in binary search it kind of equates to the case of complete tree. Degenerate trees become lists
– Jefferson Quesado
@Jeffersonquesado And that’s why they invented the binary trees with self-rebalancing. Unbalanced trees in the worst case degenerate into linked lists.
– Victor Stafusa