On hierarchical visualization of event detection in twitter

Akhtar, N. and Siddique, B. (2018) On hierarchical visualization of event detection in twitter. Advances in Intelligent Systems and Computing, 554. pp. 571-579. ISSN 21945357

Full text not available from this repository.
Official URL: https://link.springer.com/chapter/10.1007%2F978-98...

Abstract

The data generated from social networking services like Twitter, contains rich information of all kinds of events. Studies have been made for event detection events in Twitter, the focus being primarily to detect events and visually align them along a timeline. Since the events can be relatively large in number carrying unequal importance, it might be overwhelming for the user to go through all the events along the timeline. A better approach could be, if the user can get an overview of the timeline at different levels of detail and traverse to those segments in which he is more interested. In this paper, we propose a novel unified workflow in which events are detected and a hierarchy of the detected events is generated through recursive hierarchical clustering. The levels of hierarchy represent the timeline at different granularities of time. Comprehensive experiment on Twitter dataset demonstrates the effectiveness of our framework. © Springer Nature Singapore Pte Ltd. 2018.

Item Type: Article
Uncontrolled Keywords: Event detection; Hierarchical clustering; Social networking services
Depositing User: AMU Library
Date Deposited: 19 Jan 2018 03:58
Last Modified: 19 Jan 2018 03:58
URI: http://ir.amu.ac.in/id/eprint/10819

Actions (login required)

View Item View Item