Targeted advertising using temporal analysis of user specific data

• US 8,626,575 B1
• Filed: 03/20/2008
• Issued: 01/07/2014
• Est. Priority Date: 03/20/2008
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Gong, Yitao
• Primary Examiner(s): Ebersman, Bruce I
• Application Number:
  US12/052,101
• Time in Patent Office: 2,119 Days
• Field of Search: None
• US Class Current: 705/14
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0252 based on events or environment, e.g. weather or festivals
• CPC Class Codes: G06Q 30/0267 Wireless devices
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute

Abstract

Methods, systems and apparatus, including computer program products are described for providing temporally relevant targeted content items. A set of user data is used to determine current user interests as indicated by one or more recent events. Content items relevant to those events are identified and ranked based on a time that the corresponding event occurred. One or more of the top ranked content items can be delivered to a user.

Claims

1. A method to present content items, the method comprising:

identifying, by data processing apparatus, a recent event associated with a user, and a plurality of past events, each of which is associated with the user and occurred before the recent event, wherein each event of the plurality of past events includes an action by the user to add an item to a wish list before the recent event;

for each of the plurality of past events;

determining, by one or more data processing apparatus, a keyword associated with a particular past event,

determining, by one or more data processing apparatus, a temporal weight for the keyword, the temporal weight based on a time of occurrence of the particular past event relative to a time of occurrence of the recent event,

identifying, by one or more data processing apparatus, a plurality of ranked candidate content items for presentation to the user based on the keyword, and

adjusting, by one or more data processing apparatus, ranks of the plurality of ranked candidate content items based on the temporal weight for the keyword associated with the particular past event resulting in an adjusted plurality of ranked content items, wherein adjusting the ranks comprises multiplying the temporal weight for the keyword with a rank of each candidate content item of the plurality of candidate content items;

identifying a future event associated with the user, the future event scheduled to occur at a time after the recent event;

determining a keyword associated with the future event and a temporal weight associated with the keyword associated with the future event;

determining a plurality of ranked candidate content items based on the keyword associated with the future event;

adjusting ranks of the plurality of ranked candidate content items determined based on the keyword associated with the future event and the temporal weight associated with the keyword associated with the future event resulting in another adjusted plurality of ranked candidate content items;

identifying, by one or more data processing apparatus, a plurality of ranked candidate content items selected based on a keyword associated with the recent event;

and selecting, by one or more data processing apparatus, one or more highest ranked candidate content items from among the plurality of ranked candidate content items selected based on the keyword associated with the recent event, the another adjusted plurality of ranked candidate content items, and the adjusted plurality of ranked candidate content items for each of the plurality of past events for presentation to the user.

2. The method of claim 1, wherein the recent event includes a transaction between the user and a merchant, the method further comprising:

receiving a request to process a transaction between the user and a merchant; and

sending the one or more highest ranked candidate content items for display to the user near a conclusion of the transaction.

3. The method of claim 2, wherein:

the candidate content items comprise advertisements provided by merchants registered with a payment system.

4. The method of claim 1, further comprising:

displaying the one or more highest ranked candidate content items to the user.

5. The method of claim 1, wherein the temporal weight is a decimal value.

6. The method of claim 1, wherein the recent event comprises an action by the user to add an item to a wish list.

7. The method of claim 1, further comprising:

determining a temporal weight for the keyword associated with the recent event; and

adjusting ranks of the plurality of ranked candidate content items selected based on the keyword associated with the recent event, based on the temporal weight for the keyword associated with the recent event.

8. The method of claim 1, wherein:

the future event is scheduled in user data for one or more applications offered by a distributed applications provider.

9. The method of claim 1, further comprising presenting the one or more highest ranked candidate content items to the user, wherein each of the one or more highest ranked candidate content items includes a selectable hyperlink that addresses a respective highest ranked candidate content item.

10. The method of claim 1, further comprising:

filtering candidate content items, wherein filtering candidate content items includes applying at least one of a whitelist or a blacklist.

11. The method of claim 1, wherein:

the list of adjusted rank content items comprises candidate content items from at least two sets of candidate content items.

12. A system comprising:

data processing apparatus; and

a non-transitory computer-readable medium storing computer program instructions executable by the data processing apparatus to perform operations comprising;

receiving a request to display content items to a user during a recent event associated with the user at a particular time;

identifying past user events associated with the user, each of which occurred before the particular time, wherein each of the past user events includes an action by the user to add an item to a wish list before the recent event;

for each of the past user events;

determining a respective keyword;

determining a temporal weight that corresponds to the respective keyword;

generating a list of ranked candidate content items for displaying to the user based on the respective keyword; and

generating an adjusted rank list of candidate content items by applying the temporal weight that corresponds to the respective keyword to each candidate content item in the list of ranked candidate content items, wherein applying the temporal weight comprises multiplying a ranking score of each candidate content item in the list by the temporal weight determined for the respective keyword;

identifying a future event associated with the user, the future event scheduled to occur after the recent event;

determining a keyword associated with the future event and a temporal weight associated with the keyword associated with the future event;

generating a list of ranked candidate content items for displaying to the user based on the keyword determined for the future event;

generating an another adjusted rank list of candidate content items based on the keyword associated with the future event and the temporal weight associated with the keyword associated with the future event;

identifying a plurality of ranked candidate content items based on a keyword associated with the recent event, wherein the keyword is associated with a respective temporal weight;

adjusting ranks of the plurality of ranked candidate content items based on the temporal weight associated with the keyword associated with the recent event resulting in an adjusted plurality of ranked candidate content items;

selecting one or more highest ranked candidate content items from among the adjusted plurality of ranked candidate content items, the another adjusted rank list of candidate content items, and the adjusted rank list of candidate content items for each of the past user events; and

displaying the one or more highest ranked candidate content items in response to receiving the request.

13. The system of claim 12, wherein the recent event includes a transaction between a user and a merchant at the particular time.

14. The method of claim 13, wherein generating a list of ranked candidate content items for displaying to the user based on the respective keyword comprises searching advertisements provided by merchants registered with a payment system.

15. The system of claim 12, wherein receiving the request to display content items to the user comprises receiving a request to identify one or more content items to be placed on a web page of an Internet publisher for display to the user.

16. The system of claim 12, wherein:

the content items comprise advertisements.

17. The system of claim 12, wherein the past events comprise at least one of past user purchases or adding items to a wishlist.

18. A non-transitory computer-readable medium storing computer program instructions executable by data processing apparatus to perform operations comprising:

receiving an indication that a recent event associated with a user occurred at a particular time;

determining that a plurality of past events that occurred at respective past times are associated with the user, wherein each of the plurality of past events includes an action by the user to add an item to a wish list before the recent event;

for each past event of the plurality of past events;

identifying a respective ranked list of candidate content items for presenting to the user at the particular time, wherein each candidate content item is selected based on a particular past event, and

adjusting ranks of candidate content items in the respective ranked list of candidate content items based on a temporal weight associated with the particular past event resulting in an adjusted ranked list of candidate content items, wherein adjusting the ranks comprises multiplying the temporal weight associated with the particular past event with a rank of each candidate content item,

identifying a future event associated with the user, the future event scheduled to occur at a time after the recent event;

identifying an additional ranked list of candidate content items based on the future event;

adjusting ranks of the candidate content items in the additional ranked list of candidate content items based on a temporal weight associated with the future event resulting in an additional adjusted ranked list of candidate content items;

identifying another ranked list of candidate content items for presenting to the user at the particular time, wherein each candidate content item in the other ranked list of candidate content items is selected based on the recent event;

selecting one or more highest ranked candidate content items from the adjusted ranked list of candidate content items for each past event of the plurality of past events, the additional ranked list of candidate content items, and the other ranked list of candidate content items; and

providing the selected one or more highest ranked candidate content items in response to receiving the indication.

19. The non-transitory computer-readable medium of claim 18, wherein each past event is associated with one or more keywords, the operations to identify a respective ranked list of candidate items further comprising:

determining one or more keywords associated with each past event; and

identifying a plurality of content items to identify candidate content items that are also associated with the one or more keywords.

20. The non-transitory computer-readable medium of claim 19, the operations further comprising:

adjusting ranks of candidate content items in the other ranked list based on a temporal weight associated with the recent event resulting in another adjusted ranked list of candidate content items, and

selecting one or more highest ranked candidate content items from the adjusted ranked list of candidate content items for each past event of the plurality of past events, the additional ranked list of candidate content items, and the other adjusted ranked list of candidate content items.