How can I match bunch of elements to canonical products which is unknown? (Entity Resolution)
The problem is simple but solution is not. ChatGPT doesn't really give an answer.
What i want is to group the "apple" together, strawberry together in a big corpus of data.
These are also noisy and really different since there is shiny apple, blue apple etc..
And another problem is that i don't have an exact name called "apple", i want the program to find the canonical entities by itself without having an input, it is not a zero-shot thing.
What should i do?