Lexical association metric for knowledge-free extraction of phrasal terms

A system for identifying phrasal terms, the system comprising:

• a processor;
• and
• a memory,
• wherein the processor is configured to execute steps comprising:
• receiving a text having a plurality of words;
• determining a plurality of contexts, wherein a context comprises one or more words proximate to another word in the text;
• for each context, determining a first frequency based on a number of occurrences of the context within the text;
• assigning a first rank to at least one context based on the first frequency for the at least one context;
• determining multiple word-context pairs;
• for each word-context pair, determining a second frequency based on a number of occurrences of the word associated with the word-context pair being used in the context;
• assigning a second rank to at least one word-context pair based on the second frequency for the at least one word-context pair;
• determining a rank ratio for each word-context pair based on the first rank and the second rank associated with the word-context pair;
• determining a mutual rank ratio based on multiple rank ratios;
• and
• identifying a phrasal term based on the mutual rank ratio.

An article of manufacture comprising a storage device in the form of a memory or a computer readable disk, the storage device comprising programming instructions for identifying phrasal terms in a text, the programming instructions when executed causing a processing system to carry out steps comprising:

• receiving a text having a plurality of words;
• determining a plurality of context, wherein a context comprises one or more words proximate to another word in the text;
• for each context, determining a first frequency based on a number of occurrences of the context within the text;
• assigning a first rank to at least one context based on the first frequency for the at least one context;
• determining multiple word-context pairs;
• for each word-context pair, determining a second frequency based on a number of occurrences of the word associated with the word-context pair being used in the context;
• assigning a second rank to at least one word-context pair based on the second frequency for the at least one word-context pair;
• determining a rank ratio for each word-context pair based on the first rank and the second rank associated with the word-context pair;
• determining a mutual rank ratio based on multiple rank ratios;
• and
• identifying a phrasal term based on the mutual rank ratio.