| The proposed research is situated in the area of data mining. Data mining concerns the analysis of (sometimes large and complex forms of) data in order to find patterns that may be interesting or useful. While current data mining technology has made tremendous progress with respect to the size of data sets that can be analysed, standard techniques cannot handle data sets with complicated internal structure, so-called structured or relational data. Relational data mining focuses on this problem. Many different approaches of relational data mining exist, based on logic (inductive logic programming), graphs (graph mining), or relational databases (probabilistic relational models, statistical relational learning, &). Each have their strengths and weaknesses. This project aims at developing a new paradigm for data mining, one that is based on the analysis of annotated graphs. These are graphs where nodes and edges are annotated with extra information. The analysis of such graphs comprises both analysis of the graph structure and of the annotations. With this novel representation, data mining methods can be developed that strike an ideal balance between analysis of the graph structure, and analysis of the information in the annotations, and thus combine the advantages of the different approaches to relational mining that currently exist. Keywords: data mining, relational mining, graph mining, annotated graph mining, inductive logic programming. |
