Mining the Relevant Set of Heterogeneous Data in DataSpace Based on Graph Partitioning

Provided by: Binary Information Press
Topic: Data Management
Format: PDF
A large numbers of data sources in DataSpace and ubiquitous data object may has relevant with different types of data objects, therefore, finding the potential relevance of heterogeneous data has become an interesting research area for several years. The relevance in different data sources suffers from the main drawback of data heterogeneity. In this paper, the authors use a hidden data graph model to represent data objects in DataSpace. Using this model, they propose a novel algorithm to provide an easy, precise and rapid access to mine the heterogeneous data.

Find By Topic