Cannabis Lexicon

Cannabis Lexicon Search Tool

Getting Started

The cannabis lexicon is a collection of over 70,000 unique terms extracted from 70 million cannabis-related tweets. To explore and discover new terms, simply type a term in the field below and a series of similar terms will appear. The lexicon accepts single words, emoji, and hashtags.

To get started, we suggest searching the following terms:

  • joint - will return other words that are modes of consuming cannabis
  • #sourdiesel - will return other hashtags that are strains of cannabis
  • 🍁 - will return other emoji that are related to cannabis

Separate multiple search terms with commas

The Magic

We developed the cannabis lexicon by training a word2vec model on over 70 million cannabis-related tweets that we collected for about two years. The model calculates term similarity by "looking at" surrounding (or context) terms for each unique term. This approach offers a method for determining the semantic usage of terms without needing any prior knowlegdge of the terms themselves.
See a use example for this site's tools