Concept driven framework for Latent Table Discovery

Abstract: Database systems have to cater to the growing demands of the information age. The growth of the new age information retrieval powerhouses like search engines has thrown a challenge to the data management community to come up with novel mechanisms for feeding information to end users. The burgeoning use of natural language query interfaces compels system designers to present meaningful and customised information. Conventional query languages like SQL do not cater to these requirements due to syntax oriented design. Providing a semantic cover over these systems was the aim of latent table discovery focusing on semantically connecting unrelated tables that were not syntactically related by design and document the discovered knowledge. This paper throws a new direction towards improving the semantic capabilities of database systems by introducing a concept driven framework over the latent table discovery method.