Handling Large Volumes of Mined Knowledge with a Self-Reconfigurable Topology on Distributed Systems.
ABSTRACT Nowadays, massive amounts of data which are often geographically distributed and owned by different organisations, are being mined. As consequence, large volumes of knowledge is being generated. This causes the problem of efficient knowledge management in distributed data mining (DDM). The main aim of is to exploit fully the benefit of distributed data analysis while minimising the communication overhead. Existing DDM techniques perform partial analysis of local data at individual sites and then generate global models by aggregating the local results. These two steps are not independent since naive approaches to local analysis may produce incorrect and ambiguous global data models. To overcome this problem, we introduce a distributed knowledge map based on an efficient self-reconfiguration network topology to represent easily and exploit efficiently the knowledge mined in large scale distributed platforms. This will also facilitate the integration/coordination of local mining processes and existing knowledge to build global models. In this paper, we implement this knowledge map and present some preliminary results about its performance.
- [show abstract] [hide abstract]
ABSTRACT: Knowledge Discovery in Databases, is an inherently iterative process requiring human interaction. The traditional model for KDD process takes a process-centric view and does not allow interaction during actual mining. The gross granularity of the KDD process discourages application development by the non-expert users on the data mining systems.11/2001;
Conference Proceeding: Knowledge map creation and maintenance for virtual communities of practice[show abstract] [hide abstract]
ABSTRACT: This paper proposes knowledge map creation and maintenance approaches by utilizing information retrieval and data mining techniques to facilitate knowledge management in virtual communities of practice. Besides evaluating their performance using synthesized data, the generated knowledge maps for documents collected from the teachers' cyber community, SCTNet, and the master thesis repository at Taiwan's National Central Library, are evaluated by domain experts. Domain experts are asked to revise the obtained knowledge maps, and the proportion of modification is small and acceptable. Therefore, the developed approaches are suitable for support knowledge management of professional communities on the Internet.System Sciences, 2003. Proceedings of the 36th Annual Hawaii International Conference on; 02/2003
- [show abstract] [hide abstract]
ABSTRACT: In this paper we proposed a hierarchical P2P network based on a dynamic partitioning on a 1-D space. This hierarchy is created and maintained dynamically and provides a gridmiddleware (like DGET) a P2P basic functionality for resource discovery and load-balancing.This network architecture is called TreeP (Tree based P2P network architecture) and is based on atessellation of a 1-D space. We show that this topology exploits in an efficient way theheterogeneity feature of the network while limiting the overhead introduced by the overlaymaintenance. Experimental results show that this topology is highly resilient to a large number ofnetwork failures.CoRR. 01/2006; abs/cs/0608118.