A system and method for mining data from a database using relevance networks

摘要

Described are a system and method for mining data in databases to discover significant relationships among variables in the data. An association is established between each pair of variables. From the data, the strength of the each association is calculated. Correlation coefficients can determine the strength of the associations. In another embodiment, the strength of each association is computed according to mutual information. These calculated strengths are evaluated according to a predetermined criterion. All associations that satisfy the criterion are included in one or more relevance networks. Each relevance network is displayed to provide a pictorial view of the relevant relationships among variables in the data.