This article will show you how to create a RAG system that makes your data easily accessible through prompts.
A RAG system is an innovative approach to information retrieval. It uses traditional information retrieval approaches, such as vector similarity search, combined with state-of-the-art large language model technology. Combined, these technologies form a robust system that can access large amounts of information with a simple message.
My motivation for this article is my frustration with trying to find an old email. I usually have some information about the email, like who the correspondent was or vaguely what the subject of the email was. Still, I have to be more specific when doing a direct word search in Gmail, which makes finding the specific email I'm looking for a challenge. I would like to have a RAG system that allows me to ask my emails to fetch them. So if I needed an old email from my university on a topic, I could ask something like “Which technical course did I enroll in during my second year at NTNU?” A direct word search equivalent to this prompt is challenging as I need more specific information in my prompt. Instead, a RAG system could find the mail, as long as it has all the necessary data.
· Motivation
· recover data
· Data prior to the process
· Other options
· RAG implementation
∘ prepare the data
∘ Upload the LLM
∘ Use the LLM
· Evidence
∘ Test 1
∘ Test 2
· Future work
· Conclusion
The first step in creating a RAG system is to find the data you want your RAG system to use. In my case, I want to search for emails, although…
