Queries
Embodiments are directed to a computer implemented method of improving queries. The method includes performing a query on a document set, wherein documents that satisfy the query comprise a first subset of the document set and documents that do not satisfy the query comprise a complement subset. Thereafter, calculating a correlation for all terms in the complement subset, with respect to the complement subset. Thereafter determining a set of top correlated terms. A set of bottom correlated terms is determined. The set of top correlated terms and bottom correlated terms are presented to generate an improved query.
1. A computer implemented method of improving a query comprising:
performing a query on a document set, wherein documents that satisfy the query comprise a first subset of the document set and documents that do not satisfy the query comprise a complement subset;
calculating a respective correlation between the query and all terms in the complement subset, wherein the correlation is based on a likelihood that a term in the complement subset appears in the first subset;
determining a set of top correlated terms, wherein a number of the top correlated terms is a user defined number;
determining a set of bottom correlated terms, wherein a number of the bottom correlated terms is a user defined number; and
presenting the set of top correlated terms and bottom correlated terms to generate an improved query.
2. The computer implemented method of claim 1 wherein:
presenting the set of top correlated terms and bottom correlated terms comprises causing the top correlated terms and bottom correlated terms to be displayed.
3. The computer implemented method of claim 1 wherein:
presenting the set of top correlated terms and bottom correlated terms comprises determining one or more suggested query modifications; and
presenting the one or more suggested query modifications to a user.
4. The computer implemented method of claim 1 wherein:
presenting the set of top correlated terms and bottom correlated terms comprises determining one or more suggested query modifications; and
performing a query based on one of the one or more suggested query modifications.
5. The computer implemented method of claim 1 wherein:
determining the set of top correlated terms comprises:
generating a first set of terms in the complement subset;
for each term in the first set of terms, calculating a correlation between the term and the complement subset;
ranking each term of the first set of terms; and
determining the set of top correlated terms based on the ranking of each term, wherein the set of top correlated terms includes terms with a high correlation.
6. The computer implemented method of claim 5 wherein the set of top correlated terms indicates terms that are dominant in the complement subset.
7. The computer implemented method of claim 5 wherein:
determining the set of bottom correlated terms comprises:
determining the set of bottom correlated terms based on the ranking of each term, wherein the set of bottom correlated terms includes terms with a low correlation.