Realising Dynamism in MediaSense Publish/Subscribe Model for Logical-Clustering in Crowdsourcing
2014 (English)In: International Journal of Advanced Research in Artificial Intelligence, ISSN 2165-4050, Vol. 3, no 11Article in journal (Refereed) Published
The upsurge of social networks, mobile devices, Internet or Web-enabled services have enabled unprecedented level of human participation in pervasive computing which is coined as crowdsourcing. The pervasiveness of computing devices leads to a fast varying computing where it is imperative to have a model for catering the dynamic environment. The challenge of efficiently distributing context information in logical-clustering in crowdsourcing scenarios can be countered by the scalable MediaSense PubSub model. MeidaSense is a proven scalable PubSub model for static environment. However, the scalability of MediaSense as PubSub model is further challenged by its viability to adjust to the dynamic nature of crowdsourcing. Crowdsourcing does not only involve fast varying pervasive devices but also dynamic distributed and heterogeneous context information. In light of this, the paper extends the current MediaSense PubSub model which can handle dynamic logical-clustering in crowdsourcing. The results suggest that the extended MediaSense is viable for catering the dynamism nature of crowdsourcing, moreover, it is possible to predict the near-optimal subscription matching time and predict the time it takes to update (insert or delete) context-IDs along with existing published context-IDs. Furthermore, it is possible to foretell the memory usage in MediaSense PubSub model.
Place, publisher, year, edition, pages
Bradford: The Science and Information (SAI) Organization , 2014. Vol. 3, no 11
Internet, crowdsourcing; pervasive computing; context information; dynamism, context-ID; logical-clustering; Publish/Subscribe; MediaSense
Research subject Computer and Systems Sciences
IdentifiersURN: urn:nbn:se:su:diva-111093DOI: 10.14569/IJARAI.2014.031106OAI: oai:DiVA.org:su-111093DiVA: diva2:774230