Query rewriting and visualising for concept evolution

Abstract

Concept evolution in data is the process where a certain concept was applied to a type of records that evolves to a new concept that is now applied to the same type of record. There are some existing tools to discover concept evolution and query with this concept evolution in mind. Each approach does have its limitation. The application of these approaches tends to depend on the structure of the data. Concept evolution in a data set can cause problems where historical data is missed when querying because to the concept of the query underwent evolution. For these types of situation rewriting a query based on the concept evolution in the data would mitigate those problems. We introduce a system that creates a concept evolution graph based on the evolution in the data and uses it to rewrite a query to take the evolution into account. Additionally we create visualisations that help understand the evolution in the data and also specific queries.