Automatic index term augmentation in document retrieval

1. A medium storing instructions executable by at least one processor, the instructions configured to cause the at least one processor to:

create a search query comprised of at least one term in a specific document;

apply the search query to a collection of documents;

select from the collection of documents a subset of documents, the subset of documents achieving the highest scores upon application of the search query;

select at least one term for use as at least one index term for the specific document from among terms in the subset of documents based upon the co-occurrence of terms in the subset of documents with terms in the specific document.

2. The medium of claim 1 , wherein the search query comprises of all the terms in the specific document.

3. The medium of claim 1 , further storing instructions to:

assign the selected at least one term to the specific document.

4. A medium storing instructions executable by at least one processor, the instructions configured to cause the at least one processor to:

select one or more index terms from a plurality of index terms;

identify one or more documents of a plurality of documents to which each of the one or more index terms has been assigned;

compare, for each of the one or more index terms, each of the identified documents to a specific document;

determine a score for each of the one or more index terms based on the comparing; and

assign the index term associated with the highest score to the specific document.

5. A computer-implemented method comprising:

selecting one or more index terms from a plurality of index terms;

identifying one or more documents of a plurality of documents to which each of the one or more index terms has been assigned;

comparing, for each of the one or more index terms, each of the identified documents to a specific document;

determining a score for each of the one or more index terms based on the comparing;

assigning the index term associated with the highest score to the specific document;

selecting a first category;

selecting one or more second categories assigned to a first supercategory;

comparing the first category and the one or more second categories;

computing a first score for the first supercategory based on the comparisons;

selecting one or more third categories assigned to a second supercategory;

comparing the first category and the one or more third categories;

computing a second score for the second supercategory based on the comparisons;

assigning the first category to the first supercategory when the first score is higher than the second score;

assigning the first category to the second supercategory when the second score is higher than the first score;

wherein the comparing of each of the identified documents to a specific document and the comparing the first category, comparing of the first category and the one or more second categories, comparing the one or more second categories, and the assigning the index term associated with the highest score to the specific document each include computing a likelihood ratio.

6. The method of claim 5 , further comprising:

receiving a first user request;

presenting a set of categories based on the first user request, the set of categories including the first category;

receiving a second user request selecting the first category;

presenting a set of items associated with the first category;

presenting an advertisement, the advertisement associated with the first supercategory when the first category is assigned to the first supercategory and associated with the second supercategory when the first category is assigned to the second supercategory.

7. The method of claim 5 , wherein the advertisement is a banner advertisement of web site.

8. The method of claim 5 , wherein the set of items is a set of stores.

9. The method of claim 5 , wherein the likelihood ratio is a log likelihood ratio.