I am currently working on a chatbot that is beginning as an FAQ bot. Its possible to build the Knowledge Base Bot using the document like (pdf, doc & docx). If user need to info then rasa chat boat search the data in the above file not on data bases. So this scenario is possible with Rasa?
Hi @Sudhakar, thank you for your question.
One of the ways to create a knowledge-based bot is by using Rasa Pro’s Enterprise Search. Here is the detailed tutorial on how to do it: Handling FAQs with Rasa and Faiss: How to implement RAG | The Rasa Blog.
Will that work for you, or are you looking for a solution specifically for Rasa Open Source?
@m_ashurkina tanks for your quick reply. So Rasa Pro’s Enterprise is not open source. Currently we are looking the same functionality Rasa Open Source. So it is possible in Rasa Open Source?
Hi @Sudhakar, the Open Source version of Rasa doesn’t include the Enterprise Search feature that’s available in Rasa Pro. If you’re looking for similar functionality, you’d need to implement it from scratch. Hope it’s helpful.
Yeah, it’s definitely possible with Rasa. You can integrate it to search through documents like PDFs or Word files by using some custom logic. You’d need to extract text from those files first—libraries like PyPDF2
for PDFs or python-docx
for Word files work well for that. After extraction, you can pass the content to Rasa’s NLU model or use it in a custom action to pull the relevant info when users ask for it.
Thanks for you response. can you please share the any example link for the same?