Methods and apparatus for generating a data dictionary

A system including:

• a plurality of modules, each module comprising instructions retained on at least one machine-readable storage medium, that when executed by a machine perform identified operations, wherein the modules comprise:
• a candidate list generator module to generate a list of keywords from search query information, and to generate token pairs including a keyword from the list of keywords and a token, the token being a synonym of the keyword;
• a validation module to calculate a divergence value for each token pair and generate a list of validated token pairs having divergence values within a divergence threshold, the divergence value calculated using a symmetric divergence measure;
• a data dictionary module for receiving the validated token pairs as entries in a vocabulary, and for providing the token pairs in response to a search query containing a keyword corresponding to at least one entry in the vocabulary.

A tangible machine readable medium storing a set of instructions that, when executed by a machine, cause the machine to:

• receive search query inputs to search a database of data items, the database having an associated data dictionary of terms to add to the search query inputs for searching the database;
• generate a set of candidate token pairs based on the search query inputs;
• calculate a symmetric cross entropy value for candidate token pairs;
• select at least one token pair according to the symmetric cross entropy value;
• build a data dictionary using the at least one token pair;
• and
• applying the data dictionary to search queries.

A method, comprising:

• executing instructions on a computing platform for:
• transmitting search query inputs to search a database of data items, the database having an associated data dictionary of terms to add to the search query inputs for searching the database;
• receiving a set of token pairs based on the search query inputs, each token pair comprising a keyword from at least one search query input and a token, the token being a synonym of the keyword;
• receiving a bi-directional divergence measure for the each of the token pairs, the bi-directional divergence being a predictability indicator for the token pairs for use in a search query;
• evaluating the bi-directional divergence measure for each of the token pairs for satisfaction of at least one user criteria;
• confirming a token pair based on the satisfaction of the at least one user criteria;
• and
• transmitting confirmation of the confirmed token pair for application in an updated data dictionary, the data dictionary used to search a database.