Identifying query patterns and associated aggregate statistics among search queries

1. A computer-implemented method, comprising:

identifying, within a search log of private search queries and corresponding post-search activity, a plurality of non-private n-grams that satisfy a privacy criterion and one or more private n-grams that do not satisfy the privacy criterion, wherein the plurality of non-private n-grams are contained in the private search queries;

generating, from the search log, based on the plurality of non-private n-grams, a plurality of query patterns, wherein at least some of the plurality of query patterns replace the one or more private n-grams that do not satisfy the privacy criterion with one or more wildcards;

determining, from the search log, aggregate search activity statistics associated with each respective query pattern of the plurality of query patterns, wherein the aggregate search activity statistics associated with each respective query pattern of the plurality of query patterns are indicative of search activity associated with a plurality of distinct private search queries that match the respective query pattern of the plurality of query patterns, wherein the plurality of distinct private search queries are part of the private search queries in the search log;

determining that the aggregate search activity statistics associated with a given query pattern of the plurality of query patterns satisfy a performance criterion;

receiving, from a remote computing device, over one or more computer networks, a search query;

based on a determination that the received search query matches the given query pattern of the plurality of query patterns and the determination that the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns satisfied the performance criterion, ranking query suggestions or search results that are responsive to the received search query based at least in part on the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns; and

providing the ranked query suggestions or the ranked search results to the remote computing device over the one or more computer networks;

wherein the remote computing device presents one or more of the ranked query suggestions or one or more of the ranked search results in accordance with the respective rankings of the ranked query suggestions or the respective rankings of the ranked search results.

2. The computer-implemented method of claim 1 , wherein the privacy criterion is satisfied by n-grams that occur within the search log at a frequency that satisfies a frequency threshold.

3. The computer-implemented method of claim 1 , wherein the privacy criterion is satisfied by n-grams that occur within the search log at least a minimum number of times.

4. The computer-implemented method of claim 1 , wherein the generating comprises generating a histogram of the plurality of non-private n-grams based on how frequently each of the plurality of non-private n-grams occurs among the private search queries in the search log.

5. The computer-implemented method of claim 1 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns include a total number of search queries that match the given query pattern of the plurality of query patterns.

6. The computer-implemented method of claim 1 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns include a number of distinct search queries that match the given query pattern of the plurality of query patterns.

7. The computer-implemented method of claim 1 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns include aggregate statistics pertaining to post-search activity associated with a plurality of private search queries that match the given query pattern of the plurality of query patterns.

8. The computer-implemented method of claim 1 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns include click position statistics associated with the given query pattern of the plurality of query patterns.

9. The computer-implemented method of claim 1 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns include a click rate associated with the given query pattern of the plurality of query patterns.

10. The computer-implemented method of claim 1 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns include an abandonment rate associated with the given query pattern of the plurality of query patterns.

11. The computer-implemented method of claim 10 , wherein the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns satisfied the performance criterion when the abandonment rate satisfies a minimum threshold.

12. A system including memory and one or more processors operable to execute instructions stored in the memory, to:

identify, within a search log of private search queries and corresponding post-search activity, a plurality of non-private n-grams that satisfy a privacy criterion and one or more private n-grams that do not satisfy the privacy criterion, wherein the plurality of non-private n-grams are contained in the private search queries;

generate, from the search log, based on the plurality of non-private n-grams, a plurality of query patterns, wherein at least some of the plurality of query patterns replace the one or more private n-grams that do not satisfy the privacy criterion with one or more wildcards;

determine, from the search log, aggregate search activity statistics associated with each respective query pattern of the plurality of query patterns, wherein the aggregate search activity statistics associated with each respective query pattern of the plurality of query patterns are indicative of search activity associated with a plurality of distinct private search queries that match the respective query pattern of the plurality of query patterns, wherein the plurality of distinct private search queries are part of the private search queries in the search log;

determine that the aggregate search activity statistics associated with a given query pattern of the plurality of query patterns satisfy a performance criterion;

receive, from a remote computing device, over one or more computer networks, a search query;

based on a determination that the received search query matches the given query pattern of the plurality of query patterns and the determination that the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns satisfied the performance criterion, rank query suggestions or search results that are responsive to the received search query at least in part based on the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns; and

provide the ranked query suggestions or the ranked search results to the remote computing device over the one or more computer networks;

wherein the remote computing device presents one or more of the ranked query suggestions or one or more of the ranked search results in accordance with the respective rankings of the ranked query suggestions or the respective rankings of the ranked search results.

13. The system of claim 12 , wherein the privacy criterion is satisfied by n-grams that occur within the search log at a frequency that satisfies a frequency threshold, or by n-grams that occur within the search log at least a minimum number of times.

14. The system of claim 12 , wherein the privacy criterion is satisfied by n-grams that occur within the search log at least a minimum number of times.

15. The system of claim 12 , wherein the generating comprises generating a histogram of the plurality of non-private n-grams based on how frequently the plurality of non-private n-grams occur among the private search queries in the search log.

16. At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by a computing system, cause the computing system to perform operations comprising:

identifying, within a search log of private search queries and corresponding post-search activity, non-private content that satisfies a privacy criterion and private content that does not satisfy the privacy criterion, wherein the non-private content is contained in the private search queries;

generating, from the search log, based on the non-private content, a plurality of query patterns, wherein at least some of the plurality of query patterns replace the private content that does not satisfy the privacy criterion with one or more wildcards;

determining, from the search log, aggregate search activity statistics associated with each respective query pattern of the plurality of query patterns, wherein the aggregate search activity statistics associated with each respective query pattern of the plurality of query patterns are indicative of search activity associated with a plurality of distinct private search queries that match the respective query pattern of the plurality of query patterns, wherein the plurality of distinct private search queries are part of the private search queries in the search log;

determining that the aggregate search activity statistics associated with a given query pattern of the plurality of query patterns satisfy a performance criterion;

receive, from a remote computing device, over one or more computer networks, a search query;

based on a determination that the received search query matches the given query pattern of the plurality of query patterns and the determination that the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns satisfied the performance criterion, altering a methodology for generating data that is presented in response to the received search query based on the aggregate search activity statistics associated with the given query pattern of the plurality of query patterns; and

provide the data to the remote computing device over the one or more computer networks;

wherein the remote computing device presents the data pursuant to the altered methodology.