Search document generation and use to provide recommendations

• US 8,005,832 B2
• Filed: 08/27/2007
• Issued: 08/23/2011
• Est. Priority Date: 08/29/2006
• Status: Expired due to Fees

• Current Assignee: Switchbook Incorporated
• Original Assignee: Switchbook Incorporated
• Inventors: Andrieu, Joseph
• Primary Examiner(s): Mahmoudi; Tony
• Assistant Examiner(s): Truong; Dennis
• Application Number:
  US11/845,697
• Publication Number:
  US 20080059454A1
• Time in Patent Office: 1,457 Days
• Field of Search: 707/3, 707/104.1, 707/706, 707/721, 707/731, 707/737, 707/999.003, 707/999.005, 707/999.007, 1/1
• US Class Current: 707/731
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Particular embodiments provide a search service using a search document. Complex, multi-site, multi-query searches may be performed using the search document. To use a search document, a user may activate the service. The search document may then capture data for an entire search performed for the users search across various online and offline resources. One or more partial search results may be determined for a search document. Partial search results may be automatically determined based on one or more resources being visited by a user during a search. Also, partial search results may be specified by a user from pages being visited, such as a user may select items on a page to be included in the search document. The partial search results are then used to provide a recommendation to the user for the specified search.

Claims

1. A method for using a target search document, the method comprising:

determining an indication for activating the target search document;

determining one or more partial search results for the target search document as a user is performing a search by accessing a plurality of resources, the one or more partial search results being determined over multiple interactions being performed in a first search by a user;

storing the one or more partial search results in the target search document;

providing an interface to the target search document to the user, allowing the user to interact with and modify the one or more partial search results;

using the target search document including the one or more partial search results for providing a recommendation for the first search; and

providing, using a computing device, the recommendation for the first search to the user using the one or more partial search results, wherein the recommendation includes a reference to a resource determined from a subset of prior search documents that are dynamically determined from a plurality of prior search documents created, modified, and submitted by other users,

wherein the subset of prior search documents are dynamically determined by;

ranking the plurality of prior search documents based on a first relevance calculation, wherein the first relevance calculation compares the plurality of prior search documents to the target search document to determine the ranking, and

selecting the subset of prior search documents, the subset including a number of prior search documents considered to be most relevant to the target search document in the ranking of plurality of prior search documents,

wherein the recommendation of the reference to the resource is dynamically provided for the first search from one or more partial search results found in the subset of prior search documents by;

ranking, a plurality of resources contained in the one or more partial search results found in the subset of prior search documents based on a second relevance calculation, the second relevance calculation based on a calculated importance of a resource in comparison to calculated importance of other resources in the plurality of resources, and

determining the resource in the ranked plurality of resources, the resource being considered to be a highest ranked resource in the plurality of resources.

2. The method of claim 1, wherein providing the recommendation comprises:

sending a representation of the target search document to a server; and

receiving the recommendation from the server, the recommendation being determined by the server based on the one or more partial search results found in the subset of prior search documents.

3. The method of claim 1, wherein providing the recommendation comprises:

determining metadata based on the one or more partial search results found in the subset of prior search documents, the metadata associated with a first resource;

determining an agent for a second resource; and

using the agent to determine the recommendation using the metadata associated with the first resource, the recommendation automatically determined from the second resource.

4. The method of claim 1, further comprising organizing the one or more partial search results found in the target search document into one or more folders in the target search document either automatically or by user-specification.

5. The method of claim 1, wherein determining one or more partial search results comprises:

reviewing information in the one or more resources;

determining a partial search result based on a relevance to the target search document; and

storing the partial search result in the target search document.

6. The method of claim 1, wherein determining one or more partial search results for the target search document comprises:

receiving a selection from the user indicating a partial search result; and

storing the selected partial search result in the target search document.

7. The method of claim 1, wherein providing the recommendation comprises using information determined from user interaction with, or modification of, one or more partial search results in the subset of prior search documents to determine the recommendation.

8. The method of claim 1, wherein the recommendation is provided by a second user, who received a representation of the target search document.

9. The method of claim 1, wherein the one or more partial search results found in the subset of prior search documents comprise indicative partial search results and potential partial search results.

20. The method of claim 1, wherein subset of prior search documents is determined based on a similarity rating of prior search documents.

21. The method of claim 1, further comprising:

receiving the plurality of prior search documents at the computing device; and

determining, at the computing device, the recommendation at the computing device using the subset of prior search documents.

10. A method for providing recommendations, the method comprising:

using a target search document for a user to determine a recommendation, the target search document including target search information related to multiple interactions performed in a first search by a user;

determining a plurality of prior search documents created, modified, and submitted by other users, each of the plurality of prior search documents including prior search information related to multiple interactions performed in a prior search;

in response to receiving the target search information, dynamically determining a subset of prior search documents from the plurality of prior search documents using the target search information, wherein the subset of prior search documents are dynamically determined by;

ranking the plurality of prior search documents based on a first relevance calculation, wherein the first relevance calculation compares the plurality of prior search documents to the target search document to determine the ranking, and

selecting, the subset of prior search documents, the subset including a number of prior search documents considered to be most relevant to the target search document in the ranking of plurality of prior search documents;

dynamically determining, using a computing device, the recommendation of a resource for the first search using the one or more partial search results found in the subset of prior search documents by;

ranking a plurality of resources contained in the one or more partial search results found in the subset of prior search documents based on a second relevance calculation, the second relevance calculation based on a calculated importance of a resource in comparison to calculated importance of other resources in the plurality of resources, and

determining the resource in the ranked plurality of resources, the resource being considered to be a highest ranked resource in the plurality of resources,

providing the determined recommendation to the user.

11. The method of claim 10, wherein determining the subset of prior search documents comprises:

determining a similarity rating for the plurality of prior search documents; and

determining the subset of prior search documents using the similarity rating for the plurality of search documents.

12. The method of claim 11, wherein the similarity rating is determined using indicative partial search results in the target search document and indicative partial search results in a prior search document to determine a similarity rating for the prior search document.

13. The method of claim 11, further comprising using the similarity rating to rate resources found within the prior search documents.

14. The method of claim 13, wherein using the similarity rating comprises:

determining resources in a prior search document; and

rating the resources based on the similarity rating.

15. The method of claim 14, wherein the resources are found in potential search results.

16. The method of claim 14, wherein rating the resources comprises:

determining a contribution value based on the similarity rating and an indicativeness rating for each resource; and

determining a recommendation value based on contribution values for similar resources across the prior search documents.

17. The method of claim 10, further comprising:

determining partial search results for the target search document; and

removing any partial search results that are found in the recommendation determined.

18. A non-transitory computer-readable storage media including logic for execution by one or more computer processors and when executed operable to:

determine an indication for activating the target search document;

determine one or more partial search results for the target search document as a user is performing a search by accessing a plurality of resources, the one or more partial search results being determined over multiple interactions being performed in a first search by a user;

store the one or more partial search results in the target search document;

provide an interface to the target search document to the user, allowing the user to interact with and modify the one or more partial search results;

use the target search document including the one or more partial search results for providing a recommendation for the first search; and

provide the recommendation for the first search to the user using the one or more partial search results, wherein the recommendation includes a reference to a resource determined from a subset of prior search documents that are dynamically determined from a plurality of prior search documents created, modified, and submitted by other users,

wherein the logic to dynamically determine the subset of prior search documents is operable to;

rank the plurality of prior search documents based on a first relevance calculation, wherein the first relevance calculation compares the plurality of prior search documents to the target search document to determine the ranking, and

select the subset of prior search documents, the subset including a number of prior search documents considered to be most relevant to the target search document in the ranking of plurality of prior search documents,

wherein the logic to dynamically provide the recommendation of the reference to the resource for the first search from one or more partial search results found in the subset of prior search documents is operable to;

rank a plurality of resources contained in the one or more partial search results found in the subset of prior search documents based on a second relevance calculation, the second relevance calculation based on a calculated importance of a resource in comparison to calculated importance of other resources in the plurality of resources, and

determine the resource in the ranked plurality of resources, the resource being considered to be a highest ranked resource in the plurality of resources.

19. A non-transitory computer-readable storage media including logic for execution by one or more computer processors and when executed operable to:

use a target search document for a user to determine a recommendation, the target search document including target search information related to multiple interactions performed in a first search by a user;

determine a plurality of prior search documents created, modified, and submitted by other users, each of the plurality of prior search documents including prior search information related to multiple interactions performed in a prior search;

in response to receiving the target search information, dynamically determine a subset of prior search documents from the plurality of prior search documents using the target information, wherein the logic to dynamically determine the subset of prior search documents is operable to;

rank the plurality of prior search documents based on a first relevance calculation, wherein the first relevance calculation compares the plurality of prior search documents to the target search document to determine the ranking, and

select the subset of prior search documents, the subset including a number of prior search documents considered to be most relevant to the target search document in the ranking of plurality of prior search documents;

dynamically determine the recommendation of a resource for the first search using the one or more partial search results found in the subset of prior search documents, wherein the logic to dynamically determine the recommendation is operable to;

rank a plurality of resources contained in the one or more partial search results found in the subset of prior search documents based on a second relevance calculation, the second relevance calculation based on a calculated importance of a resource in comparison to calculated importance of other resources in the plurality of resources, and

determine the resource in the ranked plurality of resources, the resource being considered to he a highest ranked resource in the plurality of resources,

provide the determined recommendation to the user.