URL RESCUE BY EXECUTION OF SEARCH USING INFORMATION EXTRACTED FROM INVALID URL

• US 20150066981A1
• Filed: 11/10/2014
• Published: 03/05/2015
• Est. Priority Date: 06/24/2010
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Brundage, Michael L., Williams, Sarah E.
• Granted Patent: US 9,760,632 B2
• Time in Patent Office: Days
• US Class Current: 707/769
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06F 16/9566 URL specific, e.g. using aliases, detecting broken or misspelled links

Abstract

A server system is disclosed that is capable of providing responsive content to a user when a request for an invalid URL is received. In a preferred embodiment, the server system implements multiple rescue strategies for attempting to rescue the invalid URL. One such rescue strategy involves repairing the URL by correcting for encoding errors, such as errors introduced by some web clients. Another rescue strategy involves determining that the URL contains an obsolete or outdated item identifier (e.g., product identifier) that renders the URL invalid, and returning information regarding one or more related items to the user. Another rescue strategy involves mining the invalid URL for one or more text strings reflective of the type of content desired, and then using the one or more text strings to execute a keyword search to identify content to provide to the user. The various rescue strategies may be attempted in sequence according to a hierarchy.

Claims

1. A computer-implemented URL rescue process, said URL rescue process comprising:

by a server system that hosts a site;

receiving a Uniform Resource Locator (URL) request from a user device, said URL request specifying a URL that includes a domain name and a character string;

detecting that the URL is invalid; and

in response to detecting that the URL is invalid, executing a set of one or more URL rescue strategies to attempt to locate site content that is related to the invalid URL, said set of one or more URL rescue strategies including a search term mining based strategy that comprises;

extracting, from the character string of the invalid URL, a set of one or more search terms for executing a search of content of the site, wherein extracting the set of one or more search terms comprises identifying at least one substring of the character string, and comparing the substring to entries in a selection list;

executing a search of the site content using the set of one or more search terms, wherein the search is executed using an index of site content; and

sending results of the search to the user device in response to the URL request.

2. The URL rescue process of claim 1, further comprising sending to the user device, with the search results, a notification that a requested page was not found.

3. The URL rescue process of claim 1, wherein the selection list includes a whitelist.

4. The URL rescue process of claim 1, wherein the selection list includes a blacklist.

5. The URL rescue process of claim 1, wherein the selection list comprises a list of parameters.

6. The URL rescue process of claim 1, wherein the site hosts an electronic catalog, and the search of site content is a catalog search for catalog items associated with the invalid URL.

7. The URL rescue process of claim 1, wherein the set of one or more URL rescue strategies includes a URL repair strategy that attempts to repair the invalid URL through URL modification.

8. The URL rescue process of claim 7, wherein executing the set of one or more URL rescue strategies comprises executing the search term based mining strategy after determining that the URL repair strategy is unsuccessful.

9. The URL rescue process of claim 1, wherein detecting that the URL is invalid comprises unsuccessfully attempting to resolve the URL.

10. The URL rescue process of claim 1, wherein the set of one or more URL rescue strategies includes an additional strategy that comprises:

extracting from the invalid URL an item identifier that identifies a first catalog item;

looking up in a mapping table a second catalog item that is related to the first catalog item; and

returning to the user device, in response to the URL request, a description of the second catalog item.

11. A system that hosts a site in which content is addressable using Uniform Resource Locators (URLs), the system comprising:

a data repository that stores content of the site, including page content corresponding to specific URLs;

a server system that provides network-based access to the site content in response to URL requests from user devices, said server system programmed to implement a set of one or more URL rescue strategies for processing requests for invalid URLs, said set of one or more URL rescue strategies comprising a rescue strategy in which the server system;

extracts one or more search terms from an invalid URL using a selection list;

executes a search of the content of the site using the one or more extracted search terms and an index of site content; and

returns, to a user device that requested the invalid URL, a page containing results of the search.

12. The system of claim 11, wherein the server system is configured to include in the page an indication that the URL is invalid.

13. The system of claim 11, wherein the selection list includes a blacklist.

14. The system of claim 11, wherein the site hosts an electronic catalog, and the search of content is a catalog search for catalog items associated with the invalid URL.

15. The system of claim 11, wherein the set of one or more URL rescue strategies also includes a URL rescue strategy that attempts to repair the invalid URL through URL modification.

16. The system of claim 11, wherein the server system is programmed to attempt a plurality of alternate URL rescue strategies in a defined order.

17. Non-transitory computer storage having stored thereon executable program code that directs a server system that hosts a site to implement a Uniform Resource Locator (URL) rescue process for responding to a URL request received from a user device, said URL rescue process comprising:

determining that a URL specified in the URL request is invalid; and

at least partly in response to determining that the URL is invalid, executing a URL rescue strategy that comprises;

mining, from a character string that follows a domain name portion of the invalid URL, a set of one or more search terms for executing a search of content of the site, wherein extracting the set of one or more search terms comprises identifying at least one substring of the invalid URL and comparing the substring to entries in a selection list;

executing a search of the content of the site using the set of one or more search terms, wherein the search is executed using an index of the content of the site; and

sending results of the search to the user device in response to the URL request.

18. The non-transitory computer storage of claim 17, wherein sending the results comprises sending the results with a notification that a requested resource was not found.

19. The non-transitory computer storage of claim 17, wherein the process further comprises executing a URL repair strategy that attempts to repair the invalid URL through URL modification.

20. The non-transitory computer storage of claim 17, wherein the URL rescue strategy comprises using URL delimiters to identify the substring.