Rules-based identification of items represented on web pages

• US 7,085,736 B2
• Filed: 02/27/2001
• Issued: 08/01/2006
• Est. Priority Date: 02/27/2001
• Status: Active Grant

• Current Assignee: Alexa Internet (Amazon.com, Inc.)
• Original Assignee: Alexa Internet (Amazon.com, Inc.)
• Inventors: Keezer, Paula, Tofel, Brad
• Primary Examiner(s): Fadok, Mark
• Application Number:
  US09/794,952
• Publication Number:
  US 20020143659A1
• Time in Patent Office: 1,981 Days
• Field of Search: 705/26, 705/27, 707/103, 715/513
• US Class Current: 705/27.1
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06Q 30/0641 Shopping interfaces

Abstract

The Document Object Model (DOM) of a sampled web page is used to create a rule that extracts item-related data from web pages having a similar DOM structure to the sampled web page. In response to a user request for such a web page, the rule is retrieved from a data server based on the page'"'"'s URL, and is applied to the DOM of the web page to extract item-identifying data. The item is then identified—preferably by the data server—by matching the item-identifying data to an item in a database. Supplemental information about the item is then retrieved from the database and supplied to the user'"'"'s computer for viewing in conjunction with the requested page. In a preferred embodiment, the rule is retrieved from the data server and applied to the web page by a client application that runs in conjunction with a web browser.

Claims

1. A method of identifying an item represented on a web page retrieved by a client computer, the method comprising, on the client computer:

sending a request to a server, wherein the request specifies at least a portion of an electronic address from which the web page was retrieved;

receiving a rule in response to the request, wherein the rule is associated with a plurality of web pages that share a common web page structure, and is capable of using the common web page structure to extract information from said web pages;

applying the rule to the web page, wherein applying the rule comprises, on the client computer, traversing a structural graph representation of the web page, according to a path specified by the rule, to obtain access to a node of the structural graph representation;

extracting from the web page, based at least upon access to the node, item-identifying data descriptive of the item represented on the web page;

using the item-identifying data extracted from the web page to retrieve, over a network, supplemental information about the item; and

displaying said supplemental information on the client computer in conjunction with a display of the web page; and

wherein the supplemental information comprises information previously extracted from a second web page via application of a second rule to the second web page, said second web page including information about said item.

2. The method of claim 1, wherein applying the rule to the web page comprises evaluating whether the rule is successful, at least in part, by determining whether the path leads to a single node.

3. The method of claim 2, wherein evaluating whether the rule is successful further comprises determining whether an attribute of the node is blank.

4. The method of claim 1, wherein the rule specifies at least two paths.

5. The method of claim 1, wherein the rule comprises criteria for determining whether the rule is successful, and the step of applying the rule to the web page comprises using said criteria to evaluate whether the rule succeeds.

6. The method of claim 1, wherein the electronic address is a URL.

7. The method of claim 1, wherein the structured graph representation is a document object model representation.

8. The method of claim 1, wherein the rule is a not specific to any particular web page.

9. The method of claim 1, wherein the rule is associated with, and configured to be applied to each of a plurality of web pages of, a particular Internet domain.

10. The method of claim 1, wherein the rule is associated with, and configured to be applied to each of a plurality of web pages of, a particular web site.

11. The method of claim 1, wherein the rule is derived from a sample web page of said plurality of web pages.

12. The method of claim 1, wherein the method is performed by a client application that runs on the client computer in association with a web browser.

13. The method of claim 1, wherein the rule is one of a plurality of rules associated with a domain from which the web page was retrieved, and the method comprises retrieving the plurality of rules, and attempting to apply each of the plurality of rules to the web page to extract information therefrom.

14. The method of claim 1, wherein the rule specifies a plurality of paths, and applying the rule comprises evaluating, for each path, whether the path is suitable for use in extracting information from the web page.