METHODS AND SYSTEMS FOR ASSESSING THE QUALITY OF AN ITEM LISTING

• US 20100262601A1
• Filed: 06/01/2009
• Published: 10/14/2010
• Est. Priority Date: 04/08/2009
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Yuan, Ted Tao, Rehman, Muhammad Faisal, Dumon, Olivier G., McDonald, Ryan, Ghatare, Sanjay Pundlkrao
• Granted Patent: US 9,412,127 B2
• Time in Patent Office: Days
• US Class Current: 707/727
• 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 assessing the quality of an item listing are described. In an example embodiment, a listing quality score for an item listing is derived as a weighted sum of first and second parts. The first part represents a predicted score based on a comparison of item attributes for the item listing that are known at listing time, with item attributes of similar item listings that have historical data available for assessing their actual performance. The second part is based on one or more observed demand metrics representing actual historical performance of the item listing. The weighting factor is derived, such that over time, the emphasis shifts from the predicted to the observed score.

Claims

1. A computer-implemented method comprising:

processing a query to identify item listings that satisfy the query, each item listing associated with at least one item being offered for sale;

assigning a score to each item listing that satisfies the query, the score i) representing a measure of the likelihood that an item listing, if presented in a search results page, will result in a transaction being concluded, and ii) derived as a weighted sum of first and second parts, the first part representing a predicted score based on item attributes of the item listing known at listing time, and the second part based on an observed demand metric derived from historical data performance of the item listing; and

presenting the item listings that satisfy the query in the search results page, each item listing positioned in the search results page based on the score assigned to the item listing.

2. The computer-implemented method of claim 1, wherein the predicted score based on item attributes known at listing time is derived by comparing item attributes of an item listing satisfying the query with item attributes of similar item listings that have previously concluded transactions.

3. The computer-implemented method of claim 1, wherein the predicted scored based on item attributes known at listing time is derived based on an analysis of an item attribute specifying an offering price for an item offered via an item listing satisfying the query.

4. The computer-implemented method of claim 3, wherein the analysis includes a comparison of the offering price for the item offered via the item listing satisfying the query with historical data indicating prices at which items determined to be similar to the item offered via the item listing satisfying the query have sold.

5. The computer-implemented method of claim 1, wherein the observed metrics representing actual historical performance of the item listing are based on data generated after the listing time of the item listing satisfying the query.

6. The computer-implemented method of claim 1, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of the number of transactions concluded per the number of search impressions for the item listing.

7. The computer-implemented method of claim 1, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of transactions concluded per duration of time since the item listing was first listed.

8. The computer-implemented method of claim 1, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of the number of times an item listing is selected for viewing per the number of search impressions for the item listing.

9. The computer-implemented method of claim 1, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of the number of times an item listing is added to a watch list per the number of search impressions for the item listing.

10. The computer-implemented method of claim 1, wherein the first part representing the predicted score is weighted with a weighting factor that is a function of the number of search impressions for the item listing such that the weighting factor for the first part decreases as the number of search impressions for the item listing increases.

11. The computer-implemented method of claim 1, wherein the second part representing observed metrics is weighted with a weighting factor that is a function of the number of search impressions for the item listing such that the weighting factor for the second part increases as the number of search impressions for the item listing increases.

12. The computer-implemented method of claim 1, wherein presenting the item listings that satisfy the query in the search results page, each item listing positioned in the search results page based on the score assigned to the item listing includes presenting the item listings as a list such that the item listing assigned the highest score is presented at the top of the list.

13. A server comprising:

a search engine module to process a query to identify item listings that satisfy the query, each item listing associated with at least one item being offered for sale;

a ranking score assignment module to assign a score to each item listing that satisfies the query, the score i) representing a measure of the likelihood that an item listing, if presented in a search results page, will result in a transaction being concluded, and ii) derived as a weighted sum of first and second parts, the first part representing a predicted score based on item attributes of the item listing known at listing time, and the second part based on observed metrics representing actual historical performance of the item listing; and

a presenting the item listings that satisfy the query in the search results page, each item listing positioned in the search results page based on the score assigned to the item listing.

14. The server of claim 13 wherein the predicted score based on item attributes known at listing time is derived by comparing item attributes of an item listing satisfying the query with item attributes of similar item listings that, when presented in a search results page, have resulted in transactions.

15. The server of claim 13, wherein the predicted scored based on item attributes known at listing time is derived based on an analysis of an item attribute specifying an offering price for an item offered via an item listing satisfying the query.

16. The server of claim 15, wherein the analysis includes a comparison of the offering price for the item offered via the item listing satisfying the query with historical data indicating prices at which items determined to be similar to the item offered via the item listing satisfying the query have sold.

17. The server of claim 13, wherein the observed metrics representing actual historical performance of the item listing are based on data generated after the listing time of the item listing satisfying the query.

18. The server of claim 13, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of the number of transactions concluded per the number of search impressions for the item listing.

19. The server of claim 13, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of transactions concluded per duration of time since the item listing was first listed.

20. The server of claim 13, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of the number of times an item listing is selected for viewing per the number of search impressions for the item listing.

21. The server of claim 13, wherein one of the observed metrics representing actual historical performance of the item listing is calculated as a ratio of the number of times an item listing is added to a watch list per the number of search impressions for the item listing.

22. The server of claim 13, wherein the first part representing predicted score is weighted with a weighting factor that is a function of the number of search impressions for the item listing such that the weighting factor for the first part decreases as the number of search impressions for the item listing increases.

23. The server of claim 13, wherein the second part representing observed metrics is weighted with a weighting factor that is a function of the number of search impressions for the item listing such that the weighting factor for the second part increases as the number of search impressions for the item listing increases.

24. The server of claim 13, wherein presenting the item listings that satisfy the query in the search results page, each item listing positioned in the search results page based on the score assigned to the item listing includes presenting the item listings as a list such that the item listing assigned the highest score is presented at the top of the list.

25. A method comprising:

processing a query to identify item listings that satisfy the query;

assigning the item listing that satisfies the query to one of two groups based on a determination of whether an item listing has associated with it sufficient historical data to derive a measure of performance for the item listing such that a first group includes those item listings with sufficient historical data to derive a measure of performance for the item listing, and a second group includes those item listings without sufficient historical data to derive a measure of performance for the item listing; and

presenting item listings from the first group and the second group in a search results page.