Cognitive science
Apophenia
Perceiving meaningful connections between unrelated things. It also covers gamblers who find patterns in random results.
dérive
A recommender over concept embeddings that suggests rare concepts close to or far from your taste, learned by Rocchio relevance feedback. Many come from a list I have kept while reading for years.
How it works
Closest is the unread concept most similar to the one you are reading. Other field is the one most similar to your taste, outside the fields of these two. Far is drawn at random among the 30 concepts least similar to the nearest of the last 10 you read.
Each of the 2,028 concepts is embedded from its term and one-sentence description by Qwen3-Embedding-4B (opens in a new tab). Your taste is a Rocchio vector: the mean embedding of the concepts you picked or saved among the last 50 you read, minus half the mean of those you disliked. Until you pick, save or dislike one, Other field follows the concept you are reading instead. Options you pass over are not offered again while you read the next 10. Everything runs in your browser.