Yahoo! Inc.

Hubert M Walker

Scott Bedard

Ankarino Lara

Noel C Morrison

Stephen James Blake

Timothy Caplis

Conor John Egan

Embodiments of the present invention provide automatic systems and methods for associating objects in databases of a web site by rate-based tagging. The frequencies of users entering specific tag terms for objects stored in the databases of the web site are used to determine hard associations between objects and tag terms and between objects. When the frequencies of user tags exceed established thresholds, hard associations between objects and tag terms are established. When objects are identified or determined to have hard association with tag terms, the objects are determined to be more clearly associated with the corresponding tag terms. Therefore, they should be highlighted or featured in more prominent locations on web pages of the web site to increase users' confidence in content of the web site. To identify hard-associated objects, more weights can be assigned to the hard-associated objects, which allows them to be more likely to be selected for display in prominent locations. In addition, objects that are determined to have hard associations with tag terms can also have hard associations with one another due to the common tag terms they share. The hard association relationship between objects can be displayed through links to associated objects when an object is selected for display.

Associating objects in databases by rate-based tagging

Embodiments of the present invention provide systems and methods for automatically generating tag terms (or tags) for objects in databases of a web site. The metadata of the objects (or data) of the web site are processed and parsed to automatically generate tag terms for the corresponding objects. Information (or data, or content) downloaded from the Internet often comes with metadata, which can exist in titles, description, sources, and authors of the information, etc. The metadata of downloaded information can be process and parsed to generate tag terms for the corresponding objects. The system can automatically generate tag terms for the data, which are stored as objects in the databases, and make the data (or objects) searchable. In addition, the automatically generated tag terms allow associated data to maintain their relationship. For example, data from the same sources, same authors, or same subjects can be identified based on the common tag terms. Automatically generated tag terms enable searching and association of data (or objects) in databases in a web site.

Automated tagging of objects in databases

Embodiments of the present invention provide automatic systems and methods for creating a web site based on an emerging subject, which is also a search keyword, by analyzing search activities (or data) of Internet users of a popular Internet search engine. Analysis of search activity data of the Internet search engine can also provide search keywords that are most related to the emerging subject, and Internet sites with content relevant to the emerging subject. To create such a web site that is of interests to Internet users, the emerging subject and related search keywords mentioned above can be searched on Internet sites identified to have content relevant to the emerging subject. The search results are used to populate databases for this web site. After the web site is created and launched, users of the site can contribute to the content of the site. In addition, active users of the web site can engage in basic administration of the site, which reduces the need of intervention by paid administrators. Thus, by obtaining content from Internet sites, coupled with contributions made by users, it is possible to automatically create and maintain new custom created sites with minimal intervention by paid administrators.

Growing community of inventors

Berkeley, CA, United States of America

Hubert M Walker