Classification in Multiple Heterogeneous Database Relations:A Tuple ID Predication Approach

Relational databases are the most fashionable repository for structured data, in a relational database many relations are linked together via E-R links. Multirelational classification is the procedure of building a classifier based on data stored in multiple relations and making predictions with it. Existing approaches of Inductive Logic Programming (recently, also known as Relational Mining) have proven effective with high accuracy in multirelational classification. Unfortunately, the most of them suffer from scalability problems with regard to the number of relations present in databases. In this paper, we propose a new approach, called Tuple ID Predication, which includes a set of novel and powerful methods for multirelational classification, including 1) tuple ID propagation, an efficient and flexible method for virtually joining relations, 2) new definitions for predicates and decision-tree nodes, which involve aggregated information to provide essential statistics for classification, and 3) a selective sampling method for improving scalability with regard to the number of tuples. Based on these techniques, we propose two scalable and accurate methods for multirelational classification: Tuple ID Predication Rule, a rule-based method and Mine-Tree, a decision-tree-based method. Our comprehensive experiments on both real and synthetic data sets demonstrate the high scalability and accuracy. It is very useful in effective decision making.