It's a method of determining which words in an arbitrary collection of documents (tweets, for instance) are most important when classifying those documents.
Term-Frequency-Inverse-Document-Frequency. Assigns each word a score based on how often it appears in a document relative to how often it appears in all documents.
Term-Frequency-Inverse-Document-Frequency. Assigns each word a score based on how often it appears in a document relative to how often it appears in all documents.
https://en.wikipedia.org/wiki/Tf%E2%80%93idf