A system for parallel media processing
Parallel Computing - Special issue on applications: parallel processing and multimedia
Distributed clustering using collective principal component analysis
Knowledge and Information Systems
Smart Cameras as Embedded Systems
Computer
Database Mining: A Performance Perspective
IEEE Transactions on Knowledge and Data Engineering
Data Mining: Machine Learning, Statistics, and Databases
SSDBM '96 Proceedings of the Eighth International Conference on Scientific and Statistical Database Management
Dimensions: why do we need a new data handling architecture for sensor networks?
ACM SIGCOMM Computer Communication Review
Cache-and-query for wide area sensor databases
Proceedings of the 2003 ACM SIGMOD international conference on Management of data
Distributed video arrays for tracking, human identification, and activity analysis
ICME '03 Proceedings of the 2003 International Conference on Multimedia and Expo - Volume 1
ICME '03 Proceedings of the 2003 International Conference on Multimedia and Expo - Volume 1
Beyond average: toward sophisticated sensing with queries
IPSN'03 Proceedings of the 2nd international conference on Information processing in sensor networks
Multiresolution data integration using mobile agents in distributedsensor networks
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
The coding ecology: image coding via competition among experts
IEEE Transactions on Circuits and Systems for Video Technology
System and software architectures of distributed smart cameras
ACM Transactions on Embedded Computing Systems (TECS)
World knowledge for sensors and estimators by models and internal models
Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Proceedings of the international conference on Multimedia
Learning from others: Exchange of classification rules in intelligent distributed systems
Artificial Intelligence
Hi-index | 0.00 |
In this paper we explore decentralised approaches for gathering knowledge from sensing devices. We contrast these with centralised processes like data mining, which assume that sensors, devices, or even people contributing information to a pool, do not have a sense of the ‘whole picture’ or the goal of the data collection. Thus it is necessary for a centralised mining process to create value by sorting, coordinating, and distilling the raw information. We consider instead a situation in which the contributors are given a goal, and are given the ability to co-ordinate among themselves in such a way that each can maximise its contribution to the pool. We discuss advantages of this new approach such as scalability and communication efficiency, and explore how it may change the design of devices, communication infrastructures, and algorithms, using several projects from the Media Laboratory as illustrations.