Mining models of human activities from the web
Proceedings of the 13th international conference on World Wide Web
Reality mining: sensing complex social systems
Personal and Ubiquitous Computing
SceneMaker: automatic visualisation of screenplays
KI'09 Proceedings of the 32nd annual German conference on Advances in artificial intelligence
SceneMaker: multimodal visualisation of natural language film scripts
KES'10 Proceedings of the 14th international conference on Knowledge-based and intelligent information and engineering systems: Part IV
SceneMaker: intelligent multimodal visualization of natural language scripts
AICS'09 Proceedings of the 20th Irish conference on Artificial intelligence and cognitive science
An architecture and a prototype for querying and visualising recorded context information
CONTEXT'11 Proceedings of the 7th international and interdisciplinary conference on Modeling and using context
Hi-index | 0.00 |
Context datasets are essential not only to design and evaluate context-aware systems but also to help identify important problems. For practical and ethical reasons, collecting real-life context data is not always feasible. In this paper, we investigate the possibility to use context data extracted from screenplays in certain context-aware applications. In the absence of real context data, screenplays are inexpensive, rich and well-structured substitutes. However, narrative art and techniques naturally restrict the range of applications in which they can be used. Some television series have been showing for over ten years. We argue that this substantial source of context data could benefit the community in certain applications. As a case study, we compare a context dataset extracted from a specific screenplay with one of the major publicly available real context datasets collected on mobile phones. Results in social relationship characterization are discussed.