Most
Connect the two words through a web of related words. Fewer words wins. The words are Slovak.
Every link here is a similarity between two word embeddings: vectors a model learned from how words are used. Delta Labs builds systems on this kind of model for search, recommendations and catalogue matching.
Every word is a vector learned from how Slovak is written on the web. The percentage on each link is the cosine similarity of two words × 100, and two words link when each is among the other's closest neighbours.
Type any form of a word, with or without diacritics. The warmth bar shows how close your last word is to the target. Par is the length of the model's own shortest bridge.
Built with fastText Common Crawl vectors (CC BY-SA 3.0, Facebook AI Research) and Simplemma (MIT; Slovak data from Michal Měchura's lemmatization lists, ODbL, and Wikinflection, CC BY 4.0).
See what Delta Labs builds →