Have you always dreamed of providing knowledge to a computer so that it can answer your questions? It is finally possible and easy to do this, using python and the Haystack library. I’ll explain how to quickly implement a Question Answering system, starting from docs about your favorite TV series.
Abstract
Web app: https://huggingface.co/spaces/anakin87/who-killed-laura-palmer
Repository: https://github.com/anakin87/who-killed-laura-palmer
Have you always dreamed of providing knowledge to a computer so that it can answer your questions? In this talk, I’m going to present the open source Haystack library: an end-to-end python framework that enables you to build powerful and production-ready pipelines for different search use cases, shuch as Question Answering and semantic document search. Haystack is modular, so that you can combine the best technology from Transformers, Elasticsearch, or Milvus. I present a simple and funny use case, where I quickly and easyly build a Question Answering system, starting from a collection of documents about Twin Peaks TV series. I also provide some theoretical details to make everyone understand how this technology works and I briefly talk about the trade-offs between response quality, response speed and computational effort.
