Granted Patent B2

US 9,928,313 · App. 14/739,487 · Granted Mar 27, 2018

Abstract

Claims (73)

1. A method comprising:

generating, by a computing device, a database comprising a plurality of candidate search query suggestions, a number of the candidate search query suggestions of the plurality having associated seasonal information;

receiving, over a network by the computing device, search query user input from a user computing device;

obtaining, over the network by the computing device, temporal information representing the time the search query is input by the user;

selecting, by the computing device, one or more search query suggestions from the database using the search query user input, the obtained temporal information and the associated seasonal information for the number of candidate search query suggestions, at least a portion of the seasonal information associated with each of the one or more selected search query suggestions matching the obtained temporal information, search query suggestion selection comprising generating a list comprising the selected search query suggestions and ordering the seasonal search query suggestions included in the list based on each one's respective popularity score, the generated list further comprising a number of non-seasonal search query suggestions positioned below the seasonal search query suggestions in the list, and ordering the non-seasonal search query suggestions included in the list based on each one's respective popularity score; and

making, by the computing device, a number of the one or more selected search query suggestions available for display at the user computing device.

2. The method of claim 1 , the generating a database comprising a plurality of candidate search query suggestions further comprising:

identifying, by the computing device, a plurality of search queries from one or more query logs as seasonal search queries; and

storing, in the database for each identified seasonal search query by the computing device, the seasonal search query and associated seasonal information.

3. The method of claim 2 , the associated seasonal information comprising information identifying one or more seasonal peaks for a seasonality type.

4. The method of claim 2 , the identifying further comprising:

classifying, by the computing device for a seasonality type, a query as seasonal using logged occurrences of the query and temporal information about each occurrence of the query in the one or more query logs;

detecting, by the computing device for the seasonality type, a number of seasonal peaks for the query using the logged occurrences of the query and the temporal information about each occurrence of the query in the one or more query logs.

5. The method of claim 4 , the classifying further comprising:

generating, by the computing device for the seasonality type and the query, an original time series using the logged occurrences of the query and the temporal information about each occurrence of the query, each point in the original time series defined by an associated frequency of occurrence and a time period;

estimating, by the computing device, a seasonal time series using the original time series;

determining, by the computing device, a similarity between the query's original and seasonal time series; and

classifying, by the computing device, the query as seasonal based on a determination that the determined similarity satisfies a similarity threshold.

6. The method of claim 5 , the detecting further comprising:

determining, by the computing device, a mean and standard deviation for the estimated seasonal component time series using each point's frequency of occurrence in the estimated seasonal component time series;

determining, by the computing device, a mean for the original time series using each point's frequency of occurrence in the original time series; and

determining, by the computing device and for each point in the query's original time series, whether or not the point is a seasonal peak using the mean and standard deviation determined for the estimated seasonal component time series and the mean determined for the original time series.

7. The method of claim 6 , further comprising:

reclassifying, by the computing device, the query as non-seasonal if a number of points determined to be peaks is less than a threshold number.

8. A system comprising:

a computing device comprising a processor and a storage medium for tangibly storing thereon program logic for execution by the processor, the stored program logic comprising:

generating logic executed by the processor for generating a database comprising a plurality of candidate search query suggestions, a number of the candidate search query suggestions of the plurality having associated seasonal information;

receiving logic executed by the processor for receiving, over a network, search query user input from a user computing device;

obtaining logic executed by the processor for obtaining, over the network, temporal information representing the time the search query is input by the user;

selecting logic executed by the processor for selecting one or more search query suggestions from the database using the search query user input, the obtained temporal information and the associated seasonal information for the number of candidate search query suggestions, at least a portion of the seasonal information associated with each of the one or more selected search query suggestions matching the obtained temporal information, search query suggestion selection comprising generating a list comprising the selected search query suggestions and ordering the seasonal search query suggestions included in the list based on each one's respective popularity score, the generated list further comprising a number of non-seasonal search query suggestions positioned below the seasonal search query suggestions in the list, and ordering the non-seasonal search query suggestions included in the list based on each one's respective popularity score; and

making logic executed by the processor for making a number of the one or more selected search query suggestions available for display at the user computing device.

9. The system of claim 8 , the generating logic for generating a database comprising a plurality of candidate search query suggestions further comprising:

identifying logic executed by the processor for identifying a plurality of search queries from one or more query logs as seasonal search queries; and

storing logic executed by the processor for storing, in the database for each identified seasonal search query, the seasonal search query and associated seasonal information.

10. The system of claim 9 , the associated seasonal information comprising information identifying one or more seasonal peaks for a seasonality type.

11. The system of claim 9 , the identifying logic further comprising:

classifying logic executed by the processor for classifying, for a seasonality type, a query as seasonal using logged occurrences of the query and temporal information about each occurrence of the query in the one or more query logs;

detecting logic executed by the processor for detecting, for the seasonality type, a number of seasonal peaks for the query using the logged occurrences of the query and the temporal information about each occurrence of the query in the one or more query logs.

12. The system of claim 9 , the classifying logic further comprising:

generating logic executed by the processor for generating, for the seasonality type and the query, an original time series using the logged occurrences of the query and the temporal information about each occurrence of the query, each point in the original time series defined by an associated frequency of occurrence and a time period;

estimating logic executed by the processor for estimating a seasonal time series using the original time series;

determining logic executed by the processor for determining a similarity between the query's original and seasonal time series; and

classifying logic executed by the processor for classifying the query as seasonal based on a determination that the determined similarity satisfies a similarity threshold.

13. The system of claim 12 , the detecting logic further comprising:

determining logic executed by the processor for determining a mean and standard deviation for the estimated seasonal component time series using each point's frequency of occurrence in the estimated seasonal component time series;

determining logic executed by the processor for determining a mean for the original time series using each point's frequency of occurrence in the original time series; and

determining logic executed by the processor for determining, for each point in the query's original time series, whether or not the point is a seasonal peak using the mean and standard deviation determined for the estimated seasonal component time series and the mean determined for the original time series.

14. The system of claim 13 , the stored program logic further comprising:

reclassifying logic executed by the processor for reclassifying the query as non-seasonal if a number of points determined to be peaks is less than a threshold number.

15. A computer readable non-transitory storage medium for tangibly storing thereon computer readable instructions that when executed cause a computing device to:

generate a database comprising a plurality of candidate search query suggestions, a number of the candidate search query suggestions of the plurality having associated seasonal information;

receive, over a network, search query user input from a user computing device;

obtain, over the network, temporal information representing the time the search query is input by the user;

select one or more search query suggestions from the database using the search query user input, the obtained temporal information and the associated seasonal information for the number of candidate search query suggestions, at least a portion of the seasonal information associated with each of the one or more selected search query suggestions matching the obtained temporal information, search query suggestion selection comprising generating a list comprising the selected search query suggestions and ordering the seasonal search query suggestions included in the list based on each one's respective popularity score, the generated list further comprising a number of non-seasonal search query suggestions positioned below the seasonal search query suggestions in the list, and ordering the non-seasonal search query suggestions included in the list based on each one's respective popularity score; and

make a number of the one or more selected search query suggestions available for display at the user computing device.

16. The computer readable non-transitory storage medium of claim 15 , the computer readable instructions that when executed cause a computing device to generate a database comprising a plurality of candidate search query suggestions further comprising instructions that when executed cause the computing device to:

identify a plurality of search queries from one or more query logs as seasonal search queries; and

store, in the database for each identified seasonal search query, the seasonal search query and associated seasonal information.

17. The computer readable non-transitory storage medium of claim 16 , the associated seasonal information comprising information identifying one or more seasonal peaks for a seasonality type.

18. The computer readable non-transitory storage medium of claim 16 , the computer readable instructions that when executed cause a computing device to identify further comprising instructions that when executed cause the computing device to:

classify, for a seasonality type, a query as seasonal using logged occurrences of the query and temporal information about each occurrence of the query in the one or more query logs;

detect, for the seasonality type, a number of seasonal peaks for the query using the logged occurrences of the query and the temporal information about each occurrence of the query in the one or more query logs.

19. The computer readable non-transitory storage medium of claim 16 , the computer readable instructions that when executed cause a computing device to classify further comprising instructions that when executed cause the computing device to:

generate, for the seasonality type and the query, an original time series using the logged occurrences of the query and the temporal information about each occurrence of the query, each point in the original time series defined by an associated frequency of occurrence and a time period;

estimate a seasonal time series using the original time series;

determine a similarity between the query's original and seasonal time series; and

classify the query as seasonal based on a determination that the determined similarity satisfies a similarity threshold.

20. The computer readable non-transitory storage medium of claim 19 , the computer readable instructions that when executed cause a computing device to detect further comprising instructions that when executed cause the computing device to:

determine a mean and standard deviation for the estimated seasonal component time series using each point's frequency of occurrence in the estimated seasonal component time series;

determine a mean for the original time series using each point's frequency of occurrence in the original time series; and

determine, for each point in the query's original time series, whether or not the point is a seasonal peak using the mean and standard deviation determined for the estimated seasonal component time series and the mean determined for the original time series.

21. The computer readable non-transitory storage medium of claim 20 , the computer readable instructions further comprising computer readable instructions that when executed cause a computing device to:

reclassify the query as non-seasonal if a number of points determined to be peaks is less than a threshold number.

Assignments (6)