TY - GEN
T1 - Temporal and information flow based event detection from social text streams
AU - Zhao, Qiankun
AU - Mitra, Prasenjit
AU - Chen, Bi
PY - 2007
Y1 - 2007
N2 - Recently, social text streams (e.g., blogs, web forums, and emails) have become ubiquitous with the evolution of the web. In some sense, social text streams are sensors of the real world. Often, it is desirable to extract real world events from the social text streams. However, existing event detection research mainly focused only on the stream properties of social text streams but ignored the contextual, temporal, and social information embedded in the streams. In this paper, we propose to detect events from social text streams by exploring the content as well as the temporal, and social dimensions. We define the term event as the information flow between a group of social actors on a specific topic over a certain time period. We represent social text streams as multi-graphs, where each node represents a social actor and each edge represents the information flow between two actors. The content and temporal associations within the flow of information are embedded in the corresponding edge. Events are detected by combining text-based clustering, temporal segmentation, and information flow-based graph cuts of the dual graph of the social networks. Experiments conducted with the Enron email dataset 1 and the political blog dataset from Dailykos 2 show the proposed event detection approach outperforms the other alternatives.
AB - Recently, social text streams (e.g., blogs, web forums, and emails) have become ubiquitous with the evolution of the web. In some sense, social text streams are sensors of the real world. Often, it is desirable to extract real world events from the social text streams. However, existing event detection research mainly focused only on the stream properties of social text streams but ignored the contextual, temporal, and social information embedded in the streams. In this paper, we propose to detect events from social text streams by exploring the content as well as the temporal, and social dimensions. We define the term event as the information flow between a group of social actors on a specific topic over a certain time period. We represent social text streams as multi-graphs, where each node represents a social actor and each edge represents the information flow between two actors. The content and temporal associations within the flow of information are embedded in the corresponding edge. Events are detected by combining text-based clustering, temporal segmentation, and information flow-based graph cuts of the dual graph of the social networks. Experiments conducted with the Enron email dataset 1 and the political blog dataset from Dailykos 2 show the proposed event detection approach outperforms the other alternatives.
UR - http://www.scopus.com/inward/record.url?scp=36348992495&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=36348992495&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:36348992495
SN - 1577353234
SN - 9781577353232
T3 - Proceedings of the National Conference on Artificial Intelligence
SP - 1501
EP - 1506
BT - AAAI-07/IAAI-07 Proceedings
T2 - AAAI-07/IAAI-07 Proceedings: 22nd AAAI Conference on Artificial Intelligence and the 19th Innovative Applications of Artificial Intelligence Conference
Y2 - 22 July 2007 through 26 July 2007
ER -