I have a corpus of a domain data in form of 10-15 books pdf and some articles and my end-goal is to make a question-answering system particular to that domain. For that, I would need a dataset on Q/A which I can use on top of something like SQuAD(Stanford Question Answering Dataset) for domain-specific knowledge

My stuck point is how to convert this corpus into a usable question-answering dataset.

My current strategy is something AllenAI has been working with. A list of their research papers on it can be found here

As I understand they use a combination of Knowledge Extraction, Natural Language Understanding, and Inference to get the job done. But I cannot find any good practical implementation.

Where can I find a good resource?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.