Knowledge discovery in temporal databases

Knowledge discovery in databases is the process of applying statistical, machine learning and other techniques to conventional database systems. Our survey in knowledge discovery systems has indicated that up to date there is no knowledge discovery system to deal with temporal databases. In this paper, we first give a brief description of temporal database systems and then we present some examples to show how the ORES temporal database management system could provide the necessary functionality to infer accurate and valuable knowledge from temporal databases. In particular, we discuss three common classes of database mining problems involving classifications, associations and sequences. We give a short description of our overall framework for knowledge discovery under research. The work focuses on two areas and their integration: on one side, data mining as a technique to increase the quality of data, and on the other side, temporal databases as a technique to keep the history of data. We believe that their integration will lead to even higher quality data