For this project, we will code a question-answering engine.
The engine is composed of the following parts :
- A search engine based on semantic similarity embedding
- A nearest neighbour approximation index
- An extractive QA model
The idea is to build each of this part separately before putting them together.