Recommendations based on branding

1. A method comprising:

maintaining an index comprising predetermined brand relationships, each predetermined brand relationship comprising a first brand, a second brand, and a recommendation score between the first brand and the second brand;

maintaining a corpus containing a plurality of user queries;

expanding, using one or more hardware processors, a seed set of brands corresponding to a category in the index, the expanding comprising:

accessing the corpus containing the plurality of user queries;

evaluating user queries of the plurality of user queries that contain a disjunction of brand terms; and

identifying a new brand to add to the seed set based on the evaluating.

2. The method of claim 1 , wherein the evaluating further comprises:

creating a candidate list comprising terms that occur frequently with seed brands; and

removing terms from the candidate list that occur less than a minimum number of co-occurrences.

3. The method of claim 1 , wherein the evaluating further comprises:

creating a candidate list comprising terms that occur frequently with seed brands; and

removing terms from the candidate list that occur less than a minimum occurrence percentage.

4. The system of claim 1 , further comprising identifying at least one variation of the first brand or the second brand.

5. The system of claim 1 , further comprising determining a semantic similarity between user queries of the plurality of queries, wherein the semantic similarity is based on common terms in the user queries.

6. The method of claim 1 , wherein the first brand corresponds to a first category and the second brand corresponds to a second category.

7. The system of claim 6 , wherein the first category and the second category correspond to a same category.

8. The method of claim 1 , further comprising:

receiving an indication of a user activity performed by a user;

identifying a brand preference from the user activity;

determining a recommendation to provide to the user based on the brand preference, the determining comprising querying the index using the brand preference to determine a further brand having a highest recommendation score; and

providing the recommendation to the user.

9. The method of claim 1 , further comprising:

receiving an indication of a user activity performed by a user;

identifying a brand preference and a category from the user activity;

determining a recommendation to provide to the user based on the brand preference, the determining comprising querying the index using the brand preference; and

providing the recommendation to the user, the recommendation comprising the brand preference in a different category than the category identified from the user activity.

10. The method of claim 1 , further comprising:

mining the user queries to identify relationships between the user queries;

mapping the user queries to categories; and

generating the plurality of brand relationships by mapping the relationships between the user queries.

11. A system comprising:

a first memory to store an index comprising predetermined brand relationships, each predetermined brand relationship comprising a first brand, a second brand, and a recommendation score between the first brand and the second brand;

a second memory to store a corpus containing a plurality of user queries; and

one or more hardware processors configured to expand a seed set of brands corresponding to a category in the index, the one or more hardware processors to expand the seed set by performing operations comprising:

accessing the corpus containing the plurality of user queries;

evaluating user queries of the plurality of user queries that contain a disjunction of brand terms; and

identifying a new brand to add to the seed set based on the evaluating.

12. The system of claim 11 , wherein the evaluating further comprises:

creating a candidate list comprising terms that occur frequently with seed brands; and

removing terms from the candidate list that occur less than a minimum number of co-occurrences.

13. The system of claim 11 , wherein the evaluating further comprises:

creating a candidate list comprising terms that occur frequently with seed brands; and

removing terms from the candidate list that occur less than a minimum occurrence percentage.

14. The system of claim 11 , wherein the one or more hardware processors are further configured to perform operations comprising:

receiving an indication of a user activity performed by a user;

identifying a brand preference from the user activity;

determining a recommendation to provide to the user based on the brand preference, the determining comprising querying the index using the brand preference to determine a further brand having a highest recommendation score; and

providing the recommendation to the user.

15. The system of claim 11 , wherein the one or more hardware processors are further configured to perform operations comprising:

receiving an indication of a user activity performed by a user;

identifying a brand preference and a category from the user activity;

determining a recommendation to provide to the user based on the brand preference, the determining comprising querying the index using the brand preference; and

providing the recommendation to the user, the recommendation comprising the brand preference in a different category than the category identified from the user activity.

16. A tangible machine-readable storage device having instructions embodied thereon that, when executed by one or more hardware processors of a machine, cause the machine to perform operations comprising:

maintaining an index comprising predetermined brand relationships, each predetermined brand relationship comprising a first brand, a second brand, and a recommendation score between the first brand and the second brand;

maintaining a corpus containing a plurality of user queries;

expanding a seed set of brands corresponding to a category in the index, the expanding comprising:

accessing the corpus containing the plurality of user queries;

evaluating user queries of the plurality of user queries that contain a disjunction of brand terms; and

identifying a new brand to add to the seed set based on the evaluating.

17. The tangible machine-readable storage device of claim 16 , wherein the evaluating further comprises:

creating a candidate list comprising terms that occur frequently with seed brands; and

removing terms from the candidate list that occur less than a minimum number of co-occurrences.

18. The tangible machine-readable storage device of claim 16 , wherein the evaluating further comprises:

creating a candidate list comprising terms that occur frequently with seed brands; and

removing terms from the candidate list that occur less than a minimum occurrence percentage.

19. The tangible machine-readable storage device of claim 16 , wherein the operations further comprise:

receiving an indication of a user activity performed by a user;

identifying a brand preference from the user activity;

determining a recommendation to provide to the user based on the brand preference, the determining comprising querying the index using the brand preference to determine a further brand having a highest recommendation score; and

providing the recommendation to the user.

20. The tangible machine-readable storage device of claim 16 , wherein the operations further comprise:

receiving an indication of a user activity performed by a user;

identifying a brand preference and a category from the user activity;

determining a recommendation to provide to the user based on the brand preference, the determining comprising querying the index using the brand preference; and

providing the recommendation to the user, the recommendation comprising the brand preference in a different category than the category identified from the user activity.