This project is inspired from Allen Institute for Artificial Intelligence's (AI2) Aristo Project.
The goal is to build a framework that is capable of answering questions of the 4th grade science exam level.
The IPython notebooks describe various methodologies proposed by us to build such a Question Answering framework.
They majorly contain the implementation for following approaches:
-
Bag-of-words matcher for candidate sentence extraction + Bag-of-word matcher for answer retrieval
-
Bag-of-words matcher for candidate sentence extraction + Dependency tree matcher for answer retrieval
-
Bag-of-words matcher for candidate sentence extraction + Semantic Frame Matching (SEMAFOR) for answer retrieval
-
Semantic Frame Matching (SEMAFOR) for candidate sentence extraction + Semantic Frame Matching (SEMAFOR) for answer retrieval