Indexing of digitized entities

• US 7,539,675 B2
• Filed: 04/10/2006
• Issued: 05/26/2009
• Est. Priority Date: 03/13/2001
• Status: Expired due to Fees

• Current Assignee: Picsearch AB
• Original Assignee: Picsearch AB
• Inventors: Andersson, Nils, Risberg, Robert
• Primary Examiner(s): Wong, Don
• Assistant Examiner(s): Brown, Sheree N
• Application Number:
  US11/401,129
• Publication Number:
  US 20060190445A1
• Time in Patent Office: 1,142 Days
• Field of Search: 707/5, 707/10, 709/219, 709/226
• US Class Current: 1/1
• CPC Class Codes: G06F 16/41 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

The invention relates to indexing of digitized entities in a large and comparatively unstructured data collection, for instance the Internet, such that text-based searches with respect to the data collection can be ordered via a user client terminal. Index information is generated for each digitized entity, which contains distinctive features being ranked according to a rank parameter. The rank parameter indicates a degree of relevance of particular distinctive feature with respect to a given digitized entity and is derived from fields or tags associated with one or more copies of the digitized entity in the data collection. The index information is stored in a searchable database, which is accessible via a user client interface and a search engine. The derived distinctive features and the rank parameter thus provides a possibility to carry out text-based searches in respect of non-text digitized entities, such as images, audio files and video sequences and obtain a highly relevant search result.

Claims

1. A server/client system for searching for digitized non-text entities in a data collection comprising:

an indexing input device for collecting basic information pertaining to at least one distinctive feature and at least one locator for each digitized non-text entity in a set of non-text entities from the data collection, the indexing input device including an index generator for receiving the basic information and producing in response thereto at least one rank parameter for each digitized non-text entity,

an index database for storing index information relating to the digitized non-text entities in the set,

a search engine for receiving search directives and in response thereto performing searches in the index database, and

a user client interface for receiving a search request from at least one user client terminal, forwarding the search request as a search directive to the search engine, receiving a hit list of digitized non-text entities and returning a result of a corresponding search in the index database to the at least one user client terminal,

wherein the index database is organized such that the index information for a particular digitized non-text entity comprises the at least one rank parameter, which is indicative of a degree of relevance for at least one distinctive feature associated with said digitized non-text entity, wherein the at least one rank parameter is based on a first rank component that is generated by ranking each distinctive feature associated with each digitized non-text entity based on a relative number of occurrences of the distinctive feature in association with multiple copies of the digitized non-text entity in the data collection, and wherein the at least one rank parameter is further based on a second rank component that is generated by ranking each individual distinctive feature related to the digitized non-text entity based on a position of the distinctive feature in a descriptive field associated with the digitized non-text entity, wherein the rank parameter is a combination of the first rank component with the second rank component.

2. A server/client system according to claim 1, wherein the second rank component indicates a ranking of at least one individual distinctive feature related to the digitized non-text entity on basis of both the position of the least one individual distinctive feature in the descriptive field as well as a relevancy parameter reflecting the distinctive feature'"'"'s significance in relation to other distinctive features in a particular position in the descriptive field.

3. A server/client system according to claim 1, wherein the indexing input device includes an automatic data collector for finding relevant digitized non-text entities in the data collection and creating therefrom the set of non-text entities.

4. A server/client system according to claim 1, wherein each of the digitized non-text entities in the hit list is associated with an identifier, at least one rank parameter and at least one locator for indicating a storage location in the data collection.

5. A server/client system according to claim 4, wherein each of the digitized non-text entities in the hit list is also associated with an illustrative element to be displayed on the user client terminal together with the respective digitized non-text entity.

6. A server/client system according to claim 5, wherein the illustrative element is a thumbnail picture.

7. A server/client system according to claim 3, wherein the data collection is an inter-network and the indexing input device includes a web crawler.

8. A server/client system according to claim 1, wherein the digitized non-text entities include at least one of the file types:

an image, a video sequence, and an audio sequence.

9. A server/client system according to claim 8, wherein the digitized non-text entities are stored in at least one of the formats:

AIF, AIFC, AIFF, AU, AVI, BMP, DIVX, DOC, EPS, GIF, ICO, JPEG, JPG, MOV, MP3, MP4, MPEG, MPEG4, MPG, PDF, PNG, PPT, PS, QT, RA, RAM, RAS, SND, TIF, TIFF, VCD, WAV, XLS and XMP.

10. A server/client system according to claim 1, wherein the first and second rank components are combined according to the expression:

$\sqrt{\frac{{\left(\mathrm{\alpha <br><br>\Gamma <br><br>}\right)}^2+{\left(\mathrm{\beta <br><br>\Pi <br><br>}\right)}^2}{{\mathrm{\alpha <br><br>}}^2+{\mathrm{\beta <br><br>}}^2}}$

where Γ

represents the first rank component, Π

represents the second rank component, α

represents a first merge factor and β

represents a second merge factor.