Research on Multi-Level Association Rules Based on Geosciences Data

This paper proposes a framework of multisource geo-knowledge discovery with association rules. Taking into account spatial data exist semantic fuzziness, and the conversion between qualitative concept and quantitative description is uncertain, in our study, conceptual partition algorithm and membership grade judgment algorithm based on cloud model was used. Meanwhile, there are many correlations among concepts in the field of geoscience, and the underlying correlation also need to be found with different membership grade functions, therefore, a method of multi-level association rules mining was proposed. In order to enhance frequent item set discovery efficiency, improved FP (Frequent Pattern)-Growth algorithm was presented. The algorithm was used in the empirical research on judgment of fault property at the south of Longmen Mountains in Chengdu city of China. The empirical result shows that the improved FP-Growth model acts better in frequent item-set mining.