Determining information inter-relationships from distributed group discussions
Inventors
Browning, Randolph • Papp, Stefan Nicholas • Zimmermann, Bernhard G. • Bressler, Benjamin Ralph
Assignees
Publication Number
US-9450771-B2
Publication Date
2016-09-20
Expiration Date
2034-10-24
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Techniques are described for analyzing user-supplied information and/or predicting future aspects of additional information that will be supplied by users. The analyzed information may include distributed group discussions involving numerous users, occurring via user comments and other content items supplied to social networking site(s) and/or other sources. Analysis of user-supplied information may, for example, include determining particular topics and/or categories of interest during one or more time periods and quantifying corresponding amounts of user interest; determining trends for, relevant terms and attributes for, and inter-relationships between the topics and categories; predicting future amounts of user interest in the topics and categories; tracking user interactions with information about the topics and/or categories; and taking further actions based on the analyzed and/or predicted information.
Core Innovation
The invention provides techniques for analyzing user-supplied information, particularly distributed group discussions that occur across various platforms such as social networking sites. The system obtains a large number of user comments distributed over time and identifies topics, categories, and relevant terms by grouping content into 'comment groups.' These comment groups are associated with identified topics and further analyzed to determine their relevance to specified terms or attributes, such as geographical location or author.
The analyzed data enables the extraction of content categories and the assessment of inter-relationships between topics, categories, and terms of interest. The system supports the display of multi-dimensional graphs that visualize the relevance of topics to selected terms across axes. Through automated discovery, it also determines related content categories based on previously identified topics and enables further identification of additional categories. The approach facilitates not only categorization and visualization but also trend prediction, allowing the system to predict future user interest and activity regarding these topics and categories.
Another aspect of the invention involves providing information to enable visual determination of the relative relevance of multiple topics to multiple terms or attributes. The system is designed to support various forms of user interaction, such as the selection of terms of interest, assignment of terms to graph axes, and selection of specific topics from graph visualizations for more detailed information. Additionally, the system has mechanisms for supplying relevant promotional content and supporting advertiser-specified circumstances for content delivery, integrating both analytic and client-driven display and interaction capabilities.
Claims Coverage
The patent contains multiple independent claims, with the core inventive features focusing on the analysis and visualization of distributed group discussion data, and the generation of actionable, relevance-based, multi-dimensional graphs.
Analyzing distributed group discussions to generate topic-based comment groups
The invention involves obtaining information about distributed group discussions involving textual comments from multiple users on multiple topics, analyzing these comments to identify topics, and generating comment groups, each associated with a topic and comprising comments that mention that topic. This process enables further organization and analysis of large-scale distributed discussions.
Determining relevance of topics to multiple user-specified terms and multi-dimensional visualization
The system determines terms of interest that are distinct from identified topics and assigns each term to a multi-dimensional graph axis. For each topic, it computes a relevance score to the specified terms based on the included comments of the topic's comment group. The invention provides information to display a multi-dimensional graph with axes mapped to terms, using visual indicators for topics whose position represents their relative relevance, thus enabling visual determination of relevance relationships among topics and terms.
User-interactive selection of terms and axes for graph display
The claims cover user-driven features, including receiving user selections for terms of interest and indications for assigning specific terms to specific graph axes. The system responds by updating displays and allowing user selection of which topics are shown, thereby supporting customized visualization and exploration of relationships.
Discovery of additional content categories and inter-category relationships
The invention includes automated discovery of new content categories related to existing specified content categories, wherein each new category corresponds to a topic identified in pre-existing comment groups. Indications of related additional categories are provided, allowing dynamic expansion and navigation of topic-category structures.
Providing actionable information to support promotional content and advertiser interaction
The system supports the identification and provision of relevant promotional information based on analysis results. It enables scenarios where advertisers are notified of discovered categories or determined attribute relationships, and promotional content is presented to users based on advertiser instructions matched to analytic outputs.
In summary, the claims cover an end-to-end system for analyzing distributed group discussions, grouping and categorizing comments by topic, determining and visualizing the relevance of topics to arbitrary terms or attributes in a multi-dimensional, user-driven manner, identifying expanded content relationships, and incorporating promotional or advertiser-directed outputs based on analytic findings.
Stated Advantages
The techniques allow for efficient automated analysis of distributed group discussions to determine topics, categories, and their inter-relationships, thus making it possible to obtain, disseminate, and use such information in a timely manner, even with the distributed nature of input data.
The invention supports real-time or near-real-time visualization and trend prediction of user discussions, enabling clients and users to quickly view dynamics of interest across topics and categories.
The system enables intuitive, interactive exploration of relevance among topics and terms of interest, facilitating discovery of relationships that may not otherwise be apparent to users.
Automated discovery of additional related content categories and inter-category relationships allows for adaptive and comprehensive coverage of evolving discussion topics.
The integration of analytic outputs with promotional information provides benefits to external entities such as advertisers, allowing for context-sensitive delivery of promotions and brand analysis.
Documented Applications
Visualization of distributed group discussions in social networking contexts to analyze topic relevance and interest trends for end-users, clients, or advertisers.
Provision of real-time or near-real-time analytics and predictions regarding user engagement with topics, categories, content sources, and time-based trends.
Interactive exploration tools enabling users to specify terms or attributes of interest and view multi-dimensional graphical representations of topic relevance.
Automated brand analysis and promotional content delivery for advertisers, based on discovered relationships between topics, categories, and user journeys.
Interested in licensing this patent?