This was also my thought. I’ve been playing recently with some classic NLP approaches to build a note recommendation script for my note taking workflow. It employs fairly simple methods based on the token frequency modeling but gives promisingly good results in terms of recommended connections.
any update on this? Now that chat GPT is totally virial, i’ve been looking for something that i can ask my obsidian about stuff i already know. even a plugin that took open Ai’s APIs would be great.
Hells, I know this is a lot to ask, but just look at Notion AI.
This will be a great, difficult( and pricey) leap, but I believe this will be incredibly valuable.
Problem:
As my vault grows, it’s sometimes difficult to find quick answers to questions buried in notes. I know the answers are in there, but I apparently didn’t give enough context or keywords to make it easy to find. I know if GPT3.5 had access to all my notes, it could understand what I’m asking for, and find the note(s) that contain the answers.
Example:
I’m trying to remember what days I’m supposed to schedule certain meetings on. I know I have a note somewhere with that info, but search isn’t pulling it up quickly.
Propose Solution:
Hit a hotkey and a prompt window pops up. I type “What days are reserved for strategy meetings?”
My entire vault is previously indexed using OpenAI’s embeddings and/or LlamaIndex (formerly GPT-Index), so the plugin can quickly search the index, find relevant notes (and sections within notes), then send those snippets to OpenAI for summarization.
The plugin then returns the answer: “Strategy meetings should be scheduled on the last Friday of each month.”
The answer includes links to the source(s) where it found that information (the notes in my vault)
I’m no developer. I’ve been able to build a proof-of-concept in Javascript (thanks to Chat-GPT and a lot of Googling), but I don’t know how to turn it into an Obsidian plugin.
Anyone here find this idea valuable enough to develop?
After a while, our vaults will be filled with tens of thousands of notes, and there are more to come each day. These huge number of notes are impossible to process.
Suggested features include elementary one and advanced ones.
Elementary ones: auto attaching referenced sources(pdf, pics, graphs), adding tags, auto including relevant information, etc.
advanced ones: Auto-generating new information/content/ideas/connections based on current notes in the vault. This can be implemented by deploying machine learning methods, which was trained by existing books/courses/internet content.
Guys, Thanks for all the suggestions. There are a lot to be done in this direction. But we can start with easy one, like some context-dependent automation and an elementary intelligent suggestion algorithms that can detect what you are doing and suggest some relevant things to you real time. This suggested sth may be sth new.