Query suggestion for e-commerce sites

• US 9,858,608 B2
• Filed: 03/31/2016
• Issued: 01/02/2018
• Est. Priority Date: 07/30/2010
• Status: Expired due to Fees

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Hasan, Mohammad Al, Parikh, Nishith, Singh, Gyanit, Sundaresan, Neelakantan, Johnson, Brian Scott, Khurana, Udayan
• Primary Examiner(s): Hwa, Shyue Jiunn
• Application Number:
  US15/086,921
• Publication Number:
  US 20160217521A1
• Time in Patent Office: 642 Days
• CPC Class Codes: G06F 16/24 Querying
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0625 Directed, with specific intent or strategy

Abstract

Query suggestions are provided using a query log including a number of user sessions that comprise training data. The training data includes a sequence of a plurality of sets of queries, some of the sets of queries including query transitions followed by a purchase related event. From a cleaned and normalized query log stationary scores and transition scores of at least some of the plurality of sets are generated. A set of query suggestions is built and similarity scores are computed for at least some of the set of query suggestions to determine whether individual ones of the at least some of the set of query suggestions meet a predetermined assurance level. Those that meet the assurance level are included as elements of the set of query suggestions. The set of query suggestions is mixed and ranked according to a user behavior that is sought to be influenced.

Claims

1. A method for providing query suggestions for a user comprising:

receiving a low frequency query from a client machine;

analyzing a query log of user query sessions to identify a plurality of ordered sequences of low frequency queries including the low frequency query;

generating a set of transition scores for the plurality of ordered sequences of low frequency queries based at least in part on a maximum number of transitions between a first query of an ordered pair of the plurality of ordered sequences of low frequency queries and a second query of the pair to a purchase-related event, wherein an individual transition score comprises an accumulation of a popularity score and an accumulation of a purchase efficiency score;

determining a normalized set of transition scores based on a frequency of the first query of each of the plurality of ordered sequences of low frequency queries;

building a set of query suggestions for the low frequency query from the normalized set of transition scores;

computing a set of similarity metrics for members of the set of query suggestions;

determining whether one or more similarity metrics for individual suggestions of the set of query suggestions meet a predetermined similarity; and

providing to the client machine a list of the individual suggestions whose similarity metrics meet the predetermined similarity.

2. The method of claim 1 wherein the purchase efficiency score is a linear combination of a count of different purchase related events.

3. The method of claim 1 wherein the set of similarity metrics include at least one of edit distance based similarity, category based similarity, or query-product bi-partite graph based similarity.

4. The method of claim 1 further comprising deleting at least one query suggestion based on a predetermined probability that selection of the query suggestion would result in returning a null search results page.

5. A computer-readable hardware storage device having stored therein a set of instructions which, when executed by a machine, causes execution of operations comprising:

receiving a low frequency query from a client machine;

analyzing a query log of user query sessions to identify a plurality of ordered sequences of low frequency queries including the low frequency query;

generating a set of transition scores for the plurality of ordered sequences of low frequency queries based at least in part on a maximum number of transitions between a first query of an ordered pair of the plurality of ordered sequences of low frequency queries and a second query of the ordered pair to a purchase-related event, wherein an individual transition score comprises an accumulation of a popularity score and an accumulation of a purchase efficiency score;

determining a normalized set of transition scores based on a frequency of the first query of each of the plurality of ordered sequences of low frequency queries;

building a set of query suggestions for the low frequency query from the normalized set of transition scores;

computing a set of similarity metrics for members of the set of query suggestions;

determining whether one or more similarity metrics for individual suggestions of the set of query suggestions meet a predetermined similarity; and

providing to the client machine a list of the individual whose similarity metrics meet the predetermined similarity.

6. The computer-readable hardware storage device of claim 5 wherein the purchase efficiency score is a linear combination of a count of different purchase related events.

7. The computer-readable hardware storage device of claim 5 wherein the set of similarity metrics include at least one of edit distance based similarity, category based similarity, or query-product bipartite graph based similarity.

8. The computer-readable hardware storage device of claim 5 wherein the operations further comprise deleting at least one query suggestion based on a predetermined probability that selection of the query suggestion would result in returning a null search results page.

9. A system comprising at least one hardware processor coupled to computer storage to perform operations comprising:

receiving a low frequency query from a client machine;

analyzing a query log of user query sessions to identify a plurality of ordered sequences of low frequency queries including the low frequency query;

generating a set of transition scores for the plurality of ordered sequences of low frequency queries based at least in part on a maximum number of transitions between a first query of an ordered pair of the plurality of ordered sequences of low frequency queries and a second query of the ordered pair to a purchase-related event, wherein an individual transition score comprises an accumulation of a popularity score and an accumulation of a purchase efficiency score;

determining a normalized set of transition scores based on a frequency of the first query of each of the plurality of ordered sequences of low frequency queries;

building a set of query suggestions for the low frequency query from the normalized set of transition scores;

computing a set of similarity metrics for members of the set of query suggestions;

determining whether one or more similarity metrics for individual suggestions of the set of query suggestions meet a predetermined similarity; and

providing to the client machine a list of the individual suggestions whose similarity metrics meet the predetermined similarity.

10. The system of claim 9 wherein the purchase efficiency score is a linear combination of a count of different purchase related events.

11. The system of claim 9 wherein the operations further comprise deleting at least one query suggestion based on a predetermined probability that selection of the query suggestion would result in returning a null search results page.

12. The system of claim 9 wherein the set of similarity metrics include at least one of edit distance based similarity, category based similarity, or query-product bipartite graph based similarity.