Methods and systems for deriving demand metrics used in ordering item listings presented in a search results page

• US 8,370,336 B2
• Filed: 06/01/2009
• Issued: 02/05/2013
• Est. Priority Date: 04/08/2009
• Status: Active Grant

• Current Assignee: PayPal, Inc. (PayPal Holdings, Inc.)
• Original Assignee: eBay Inc.
• Inventors: Dumon, Olivier G., McDonald, Ryan, Rehman, Muhammad Faisal, Netzloff, Julie, Sun, Ken
• Primary Examiner(s): BOCCIO, VINCENT F
• Application Number:
  US12/476,134
• Publication Number:
  US 20100262596A1
• Time in Patent Office: 1,345 Days
• Field of Search: 707/723, 707/725
• US Class Current: 707/723
• CPC Class Codes: G06F 16/90335 Query processing
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0603 Catalogue ordering
• CPC Class Codes: G06Q 30/0623 Item investigation
• CPC Class Codes: G06Q 30/0629 for generating comparisons
• CPC Class Codes: G06Q 30/0631 Item recommendations

Abstract

Methods and systems for using a half-life formula for deriving demand metrics used in ordering item listings, when presenting those item listings in a search results page, are described. In some embodiments, a demand metric for an item listing is derived by monitoring events associated with item listings, such as, keeping a count of the number of search impressions an item listing receives. A half-life formula is used to ensure that events occurring earlier in time contribute less to the demand metric than more recently occurring events. The demand metric are used to order the item listings, when the item listings are being presented in a search results page.

Claims

1. A computer-implemented method comprising:

processing a search query to identify item listings satisfying the search query, each item listing presenting an item that has been assigned to at least one category and on which a bid can be placed;

deriving for each item listing satisfying the search query a demand metric based on the number of bids that have been placed on an item offered via an item listing, the demand metric derived using a half life formula such that a value for a bid contributing to the demand metric is reduced over time, as determined by the half life formula, wherein the half life formula has a configurable parameter representing the number of days until the value of an event contributing to the demand metric is reduced by half and the configurable parameter is configurable on a per item category basis; and

presenting the item listings satisfying the search query in a search results page ordered at least in part based on the corresponding demand metric for each item listing.

2. The computer-implemented method of claim 1, wherein an event contributing to the demand metric is selected from the group:

a search impression, a view, a transaction, and a watch list; and

the demand metric is derived using a half life formula such that a value for an event contributing to the demand metric is reduced over time, as determined by the half life formula.

3. The computer-implemented method of claim 1, wherein presenting the item listings satisfying the search query in a search results page ordered at least in part based on the corresponding demand metric for each item listing includes generating a list of the item listings satisfying the search query ordered based in part on the demand metric assigned to each item listing such that the item listing assigned the highest demand metric is first in the list and the item listing assigned the lowest demand metric is last in the list.

4. The computer-implemented method of claim 1, wherein the demand metric is used in calculating a ranking score assigned to each item listing satisfying the query, the ranking score used to order the item listings when presenting the item listings in the search results page.

5. A system comprising:

a hardware-implemented item listing presentation management module configured to i) process a search query to identify item listings satisfying the search query, each item listing presenting an item that has been assigned to at least one item category and on which a bid can be placed, ii) derive for each item listing satisfying the search query a demand metric based on the number of bids that have been placed on an item offered via an item listing, the demand metric derived using a half life formula such that a value for a bid contributing to the demand metric is reduced over time, as determined by the half life formula, wherein the half life formula has a configurable parameter representing the number of days until the value of an event contributing to the demand metric is reduced by half and the configurable parameter is configurable on a per item category basis, and iii) present the item listings satisfying the search query in a search results page ordered at least in part based on the corresponding demand metric for each item listing.

6. The system of claim 5, wherein an event contributing to the demand metric is selected from the group:

a search impression, a view, a transaction, and a watch list; and

the demand metric is derived using a half life formula such that a value for an event contributing to the demand metric is reduced over time, as determined by the half life formula.

7. The system of claim 5, wherein the hardware-implemented item listing presentation management module is further configured to generate a list of the item listings satisfying the search query ordered based in part on the demand metric assigned to each item listing such that the item listing assigned the highest demand metric is first in the list and the item listing assigned the lowest demand metric is last in the list.

8. The system of claim 5, wherein the demand metric is used in calculating a ranking score assigned to each item listing satisfying the query, the ranking score used to order the item listings when presenting the item listings in the search results page.

9. A server comprising:

a memory storing instructions executable by a processor, the processor configured to execute the instructions causing the server to perform a method comprising;

processing a search query to identify item listings satisfying the search query, each item listing presenting an item that has been assigned to at least one item category and on which a bid can be placed;

deriving for each item listing satisfying the search query a demand metric using a half life formula such that a value for an event contributing to the demand metric is reduced over time, as determined by the half life formula, wherein the half life formula has a configurable parameter representing the number of days until the value of an event contributing to the demand metric is reduced by half and the configurable parameter is configurable on a per item category basis; and

presenting the item listings satisfying the search query in a search results page ordered at least in part based on the corresponding demand metric for each item listing.

10. The server of claim 9, wherein an event contributing to the demand metric is selected from the group:

a search impression, a view, a bid, a transaction, and a watch list.

11. The server of claim 9, wherein presenting the item listings satisfying the search query in a search results page ordered at least in part based on the corresponding demand metric for each item listing includes generating a list of the item listings satisfying the search query ordered based in part on the demand metric assigned to each item listing such that the item listing assigned the highest demand metric is first in the list and the item listing assigned the lowest demand metric is last in the list.

12. The server of claim 9, wherein the demand metric is used in calculating a ranking score assigned to each item listing satisfying the query, the ranking score used to order the item listings when presenting the item listings in the search results page.

13. A computer-implemented method comprising:

processing a search query to identify item listings satisfying the search query, each item listing including information concerning an item being offered for sale at a fixed price or by auction, and each item assigned to at least one item category;

deriving for an item listing satisfying the search query a demand metric based on a number of user-events of different types associated with the item listing, the demand metric derived using a half life formula such that a value for any particular event contributing to the demand metric is reduced over time, as determined by the half life formula, wherein the half life formula has a configurable parameter representing the number of days until the value of an event contributing to the demand metric is reduced by half and the configurable parameter is configurable on a per item category basis; and

presenting the item listings satisfying the search query in a search results page ordered at least in part based on the corresponding demand metric for each item listing.

14. The computer-implemented method of claim 13, wherein each different type of event that contributes to the demand metric has a separate half life formula with a different configurable parameter representing the number of days until the value of an event of the particular type contributing to the demand metric is reduced by half.