In this project, we were asked to create a natural language pipeline that would power a chatbot for open domain question answering.  A (hypothetical) restaurant chain has much text-based data on their website with their menu, history, location, hours, and other information and they would like the added ability for a website visitor to ask a question in a query box and our deep learning NLP chatbot to find the relevant information and present that back.  

We got our started showing how we could build a simple FAQ chatbot that took in random queries and matched that up to predefined questions and output a response with a confidence score that indicated the similarity between the input question and the question in our database. But this was only a stepping stone to our real goal which was to create a chatbot that could capture the intent of the question and prepare an appropriate response.

We explored a Named Entity Recognition (NER) approach to give us the added power we needed to quickly classify input text that we could then match to the relevant content for a response.  This was determined to fit our goal of allowing for open domain question answering and to take advantage of a large corpus of unstructured data that changes without using hardcoded heuristics (as in our hypothetical restaurant example).

We learned to use the building blocks of the NLP model like pre-process, tokenize, and tagging parts of speech.  We use this understanding to build a system able to read an unstructured text, to comprehend an answer to a specific question. Specifically, we gained these skills in this project:

1. How to build a basic FAQ based chatbot using statistical modeling in a framework capable of detecting intents and entities for Open-domain question-answering.
2.  Learned to generate a dense representation of sentences.
3. How to build a Document Reader for extracting answers from unstructured text.
4. Learned how to integrate deep learning models into a classic NLP pipeline.

These skills will come in very handy in your career as you see similar business use cases and as conversational user interfaces continue to gain in popularity. Well done and let's see what's in store for our next Deep Learning Projects in Python!