{A Matter of Time: how to visually analyze multivariate (and multidimensional) data, irregularly sampled and having multiple granularities}

Conference Paper
Multi Relational Data Mining searches for patterns that involve multiple tables from a relational database. In order to avoid the generation of a huge relation involving all of the attributes and the loss of information, including essential semantic information represented by the links in the database design, it aims to discover knowledge directly from relational data. Time is an intrinsic data dimension in many domains, which is different from any other dimension. However, the complexity of dealing with many temporal data at a time and to contemporary provide a visual aid to users, represents a challenge still open. This work represents an attempt to exploit the ideas coming from Multi Relational Data Mining (and Temporal Data Mining) in order to be able to analyze multivariate and multigranular time-oriented data, and to discover patterns, trends and relations among data previously unknown. Moreover it presents the preliminary steps performed so far and the steps to be performed to present, interact and communicate visual overviews of the available data and the results of the performed analysis.
Year of Publication
Conference Name
Doctoral Consortium in AIME 07 Conference
Download citation