Temporal link analysis of linked entities

• US 7,792,827 B2
• Filed: 12/31/2002
• Issued: 09/07/2010
• Est. Priority Date: 12/31/2002
• Status: Expired due to Fees

• Current Assignee: LinkedIn Corporation (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Weiss, Uri, Amitay, Einat, Herscovici, Michael, Soffer, Aya, Carmel, David, Lempel, Ronny
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): DANG, THANH HA T
• Application Number:
  US10/335,357
• Publication Number:
  US 20040128273A1
• Time in Patent Office: 2,807 Days
• Field of Search: 707/10, 707/103, 707/104.1, 707/999.001, 707/999.01, 707/999.107, 709/217, 709/218, 714/205, 714/208
• US Class Current: 707/726
• CPC Class Codes: G06F 16/972 Access to data in other repository systems, e.g. legacy data or dynamic Web page generation

Abstract

A method for gathering and recording temporal information for a linked entity, the method including identifying a link related activity within a linked source entity, and recording a time stamp in association with the link related activity.

Claims

1. A method for analyzing linked entities, comprising:

identifying one or more destination documents and a collection of source documents that contain hypertext links that identify one or more of the destination documents by specifying respective locations of the destination documents;

detecting changes in the source documents;

creating, on a computer, link activity records responsively to the detected changes, each link activity record comprising;

an identification of a respective destination document and of a respective source document in which a change was made and which contains a link to the respective destination document; and

a time stamp indicative of a time at which the change was made; and

collecting the link activity records with respect to the links from a plurality of the source documents to a given destination document, and processing the collected link activity records in order to discover changes in an activity level indicated by a number of the link activity records associated with the collection over time.

2. The method according to claim 1, wherein detecting the changes comprises detecting an addition of a link in one of the source documents.

3. The method according to claim 1, wherein detecting the changes comprises detecting a modification of anchor text associated with a link in one of the source documents.

4. The method according to claim 1, wherein detecting the changes comprises detecting that a location of link in one of the source documents has changed.

5. The method according to claim 1, wherein detecting the changes comprises detecting a deletion of a link in one of the source documents.

6. The method according to claim 1, wherein detecting the changes comprises crawling over a document collection in order to identify and detect the changes in the source documents.

7. The method according to claim 1, wherein processing the collected link activity records comprises sorting the link activity records according to the time stamp.

8. The method according to claim 7, wherein processing the collected link activity records comprises projecting an indication of the activity level onto a time axis.

9. The method according to claim 8, wherein projecting the indication comprises measuring a respective number of the link activity records occurring in each of a sequence of time intervals on a time axis.

10. The method according to claim 9, wherein processing the collected link activity records comprises identifying a trend in the activity level with respect to one or more of the documents.

11. The method according to claim 9, wherein processing the collected link activity records comprises comparing the respective number of the link activity records that are associated with the links to each of two or more of the destination documents in the sequence of the time intervals in order to detect an abnormal change in the activity level of at least one of the destination documents.

12. The method according to claim 7, wherein processing the collected link activity records comprises identifying and displaying a most recent respective link activity record for each of a plurality of the links.

13. The method according to claim 1, wherein processing the collected link activity records comprises assigning respective weights to the links responsively to the link activity records, and ranking a group of the documents according to the weights of the links that are respectively associated with the documents.

14. The method according to claim 13, wherein assigning the respective weights comprises assigning a respective weight to each link responsively to the time stamp of at least one of the link activity records that is associated with the link.

15. The method according to claim 14, wherein assigning the respective weight comprises assigning greater weight responsively to recent link activity records that are indicative of addition or modification of the links than to older link activity records.

16. The method according to claim 14, wherein assigning the respective weight comprises reducing the respective weight of a given link responsively to a link activity record indicative of a deletion of the given link.

17. The method according to claim 13, wherein the group of the documents comprises a plurality of the destination documents, and wherein ranking the group of the documents comprises aggregating the respective weights of the links to each of the destination documents, and ranking the destination documents according to the aggregated weights.

18. The method according to claim 13, wherein assigning the respective weights comprises assigning a respective weight to each link responsively to a number of the link activity records that are associated with the link.