Combining heterogeneous classifiers for relational databases

Practical usage of machine learning is gaining strategic importance in enterprises looking for business intelligence. However, most enterprise data is distributed in multiple relational databases with expert-designed schema. Using traditional single-table machine learning techniques over such data not only incur a computational penalty for converting to a flat form (mega-join), even the human-specified semantic information present in the relations is lost. In this paper, we present a practical, two-phase hierarchical meta-classification algorithm for relational databases with a semantic divide and conquer approach. We propose a recursive, prediction aggregation technique over heterogeneous classifiers applied on individual database tables. The proposed algorithm was evaluated on three diverse datasets, namely TPCH, PKDD and UCI benchmarks and showed considerable reduction in classification time without any loss of prediction accuracy.

► This paper proposes a practical, heterogeneous classifier for relational databases.
► The algorithm works directly on real databases without requiring fusion.
► The implicit semantics hidden in the database schema is used for better performance.
► A new recursive aggregation technique using a concept of Join Graph has been proposed.
► A simple proof based on structural induction has been provided to validate the proposed technique.