Introducing Harriet.ai

Welcome to Questions.ai, where Harriet the AI, uses Natural Language Processing to answer questions, given a query.

Harriet, a small sister to IBM's Watson, will perform two tasks: document retrieval and passage retrieval while having access to a corpus of text documents. When presented with a query (a question in English asked by the user), document retrieval will first identify which document(s) are most relevant to the query. Once the top documents are found, the top document(s) will be subdivided into passages (in this case, sentences) so that the most relevant passage to the question can be determined.

Harriet uses tf-idf to rank documents based both on term frequency for words in the query as well as inverse document frequency for words in the query. Once we’ve found the most relevant documents, there many possible metrics for scoring passages, but we’ll use a combination of inverse document frequency and a query term density measure.

Hope you enjoy the demo!

After a few AI projects, I wanted to dwell into the world of Natural Language Processing as NLP is a new technology, that I believe, would be an amazing asset to have on every piece of technology in our household devices or technology in general.

I wanted to learn whatever skills I had in Python to make a QnA AI that was similar to Watson who is a big scale and much more smarter version of Harriet. I say big scale as a question answering system of Watson’s accuracy requires enormous complexity and vast amounts of data,

Hence, out of curiosity and learning different skills in AI, I entered into the realm of NLP through this project and another project named Parser.ai (Gary).