Unlike traditional search engines that require users to know exactly what they’re looking for, this discovery engine caters to users who are venturing into the unknown - essentially guiding them through what they don’t know they don’t know.
The project is centered on developing a “Discovery Engine”, a novel concept derived from 2020's CoronaWhy.org experiments, aimed at revolutionizing how we uncover and interact with information. We couldn't do this for COVID back then since we didn't have LLMs to do magic, now we can!
Unlike traditional search engines that require users to know exactly what they’re looking for, this discovery engine caters to users who are venturing into the unknown - essentially guiding them through what they don’t know they don’t know.
you know what to search for, like “LLM and biomedical data”
you don’t know what you search for and discover what’s relevant to you
.png)
By extracting entities and their relationships from research papers to build comprehensive knowledge graphs, and then merging these into a singular, supergraph, the engine offers a unique visual exploration tool. This tool not only clusters graph nodes for simplified navigation but also integrates an AI agent to act as a knowledge synthesizer. As users navigate through the interconnected nodes, they are presented with the opportunity to discover and synthesize knowledge in areas they may not have intentionally sought out, thereby broadening their horizon of understanding and innovation.
-
