Foundations of statistical natural language processing
Foundations of statistical natural language processing
Introduction to topic detection and tracking
Topic detection and tracking
Analyzing feature trajectories for event detection
SIGIR '07 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
Using the Semantic Web for linking and reusing data across Web 2.0 communities
Web Semantics: Science, Services and Agents on the World Wide Web
From whence does your authority come?: utilizing community relevance in ranking
AAAI'07 Proceedings of the 22nd national conference on Artificial intelligence - Volume 2
Global models of document structure using latent permutations
NAACL '09 Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics
Hi-index | 0.00 |
The rapid evolution and growth of social media software has enabled hundreds of millions to interact within on-line communities on a global scale. While they enable communication through a common set of metaphors, such as discussion threads and quoting text in replies, this software uses a variety of diverging ways of representing discussion. Since the meaning of a conversation is defined not only by the content of a piece of text, but also by the relationships between pieces of text, part of the meaning of the discussion is obscured from automated processing. Search engines, which act as gateways to outsiders into the social text in a community, are reduced to giving an incomplete picture. This paper proposes a model for representing both the content and the structure of social text in a consistent way, enabling automated processing of the structure of the discussion along with its text content. It also describes a method for indexing text that uses this structural information to provide meaningful contexts for paragraphs of interest. It then describes a method for clustering text content into topic groups, using this indexing method, and also using the social structure to make informed decisions about which pieces of text to compare meaningfully.