This application claims the benefit as a Continuation of application Ser. No. 12/861,936, filed Aug. 24, 2010, the entire contents of which is hereby incorporated by reference as if fully set forth herein, under 35 U.S.C. §120. The applicant hereby rescinds any disclaimer of claim scope in the parent application or the prosecution history thereof and advises the USPTO that the claims in this application may be broader than any claim in the parent application.
FIELD OF THE INVENTION
- Top of Page
The present invention relates to content recommendation systems.
- Top of Page
The amount of content available on the Internet is growing at a staggering rate. News stories, multimedia presentations, blog entries, music, user generated content, and other forms of information are generated by a large number of sources, and there is no sign that this trend is slowing. Web sites and other publishing tools have made it trivial for authors to place content on the Internet for all to see, even if these authors do not consider themselves “tech-savvy.” At the same time, traditional news sources such as newspapers and magazines are generating more content than ever before and posting it to web sites in order to capture a larger market.
Many web sites, such as the “My Yahoo!®,” page by Yahoo! Inc., allow users to view original content from the web site provider, as well as content from a variety of partners and other third-party sources. By using content feeds such as RSS (Really Simple Syndication) feeds, Atom feeds, and other content distribution technologies, users are able to select content sources and customize their home page so that it reflects their interests. For example, one user may place the “Oddly Enough News” feed by Thompson Reuters on their home page alongside other news feeds, video feeds, blog feeds, and other content. Another user may choose an entirely different set of content sources to follow. News feed aggregators and other content selection technologies also allow similar functionality, with varying degrees of control, features, and presentation options.
Even though a “human filter” is constructively applied to the available content through the content selection process, it is common for users to profess interest in a much larger amount of information than they are capable of digesting. For example, users may subscribe to scores of news feeds and other groups of content, but may only have the time or desire to read, watch, or otherwise consume a small percentage of that content.
A certain amount of the content delivered to users is of little or no interest or value to the user, even if the user has time to digest the content. This is true even of content delivered via content sources selected by the user. For example, a user that has selected a content feed named “Politics” may only be interested in one or two of the stories presented via that feed. The lack of interest may, among other reasons, be due to a feed selection mismatch, or may be a symptom of a quality problem with the feed logic that selects content for that feed.
Since only a small amount of content delivered by each content source is consumed by users, it is common for users to select multiple feeds that are similar to one another. For example, a user may select several content feeds from publishers of national news or aggregators of political news. Although this technique may increase the number of content items, e.g., news stories, articles, and other media, it may also increase clutter, making it difficult for users to select relevant content without being overloaded by irrelevant content.
The relevancy of the content may vary from content provider to content provider. One content provider may take pride in matching the description of the feed with the contents provided via the feed, and may use advanced methods to ensure maximum relevancy to the broadest consumer base. However, relevancy may also vary from user to user. In other words, content that is interesting to one user may be uninteresting to another based on mood, time of day, level of education, or many other “human” factors.
Facebook®, a social networking web site, has circumvented the content aggregation model described above by allowing users who are friends with one another, i.e., connected via the web site, to recommend content to each other. Users post links to news articles, videos, and other content items that are accessible via the Internet, sometimes along with introductory comments meant to introduce the content to friends. Comments may be attached to postings by other users of the web site, such as friends of the user. In addition, friends of the posting user may signal their approval of content by pressing a “thumbs up” button. This content, however, is found in a sea of personal postings and micro blog updates by friends, and users are unable to specify content categories that they are interested in.
CNN® uses information gathered from Facebook® to list stories that are popular on the social networking site. For example, a large number of users may post links to CNN® stories, and CNN® provides a list of the most popular stories to users of the CNN® web site. In addition, Facebook® provides “social plug-ins” that allow users of web sites such as the CNN® web site, to view a list of content items from the same web site that are recommended by the user's friends on Facebook®. However, the user must be currently visiting the CNN® web site, and will only see CNN® web site stories that are linked to or recommended by friends.
The approaches described in this section are approaches that could be pursued, but not necessarily approaches that have been previously conceived or pursued. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section.
BRIEF DESCRIPTION OF THE DRAWINGS
- Top of Page
The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which like reference numerals refer to similar elements and in which:
FIG. 1A illustrates a block diagram of an environment in which an embodiment may be implemented.
FIG. 1B illustrates a block diagram of an environment in which an embodiment may be implemented.
FIG. 1C illustrates a block diagram of an environment in which an embodiment may be implemented.
FIG. 2 illustrates a logical diagram representing a social graph and social graph elements in an embodiment.
FIG. 3 illustrates sample output generated in an embodiment.
FIG. 4 illustrates a method for aggregating content in an embodiment.
FIG. 5 illustrates a method for generating a web page using social graph information in an embodiment.
FIG. 6 illustrates a computer system upon which an embodiment may be implemented.
- Top of Page
In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, that the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to avoid unnecessarily obscuring the present invention.
Users of the Internet often “register,” or create “accounts” with a variety of web sites by generating unique identifiers such as user names. By registering with a web site, a user becomes uniquely identifiable to the web site, allowing the user to create, request and receive personalized content. Users often must log in to each web site in order to view this personalized content.
According to one embodiment, a user may log in to a personalized home page, which is configured to deliver personalized news feeds and other content to the user. Multiple content feeds such as RSS feeds, each coming from a different content provider, are requested by the home page provider. These content feeds may have been chosen by the user when configuring the personalized home page. For example, a first news feed may be retrieved from CNN®, and a second news feed may be retrieved from Fox News®. Each of the news feeds includes a list of content objects that identify news articles. The content objects may include, for example, a title for the news article and a link to that news article. A small portion of the article, such as the first three lines of text, may also be included in the content object.
In an embodiment, the user is also a member of a social networking website, where the user is connected to other users. These other users are often referred to as “friends” or “connections” of the user. Friends of the user may have already viewed some of the news articles and provided feedback on them. For example, one friend may have posted a link to one of the articles on the social networking website. Another friend may have provided a rating for the article, giving it four out of five stars. These actions, and other actions associated with content are called “social signals.” Evidence of social signals may be stored by social networking providers.