LexFreq

LexFreq is a machine learning tool created by Tuğrul Ertürk to establish semantic relationships between words. It uses BNC database and word frequency lists compiled by Adam Kilgarriff (RIP).

Start the tool

How does it work?

  • There are 6318 entries in the frequency list. LexFreq randomly picks a word from the list and shows the unselected words (You can also search for and select a word).
  • If a user selects one or more semantically related word from the list, LexFreq marks them as "Has Relative Within Range."
  • While the experiment is running, LexFreq creates a relation between the words and guesses new relationships with high accuracy.
  • You can view the results at any given time by using search and the individual word's detail page.
  • There is a decontamination option to renew the samples (Because we rely on the human eye to determine the relationship).
  • The quote on the title belongs to Emil Cioran.