Research >> Projects

Knowledge Discovery in Databases

Back to List

Knowledge Discovery in Databases (KDD) has been ranked as one of the most important

research topics in 1990s by both database and machine learning researchers. In essence, KDD is the process of nontrivial extraction of implicit, previously unknown, and potentially useful information from databases.

At present the majority of existing KDD approaches rely on statistical and artificial intelligence techniques. The role of databases in KDD is limited to providing efficient access to data. It is difficult for many of existing KDD systems to scale up to massive amounts of data which change and grow quickly in real-world databases. As such, we argue that databases should play a more active role in knowledge discovery to help overcome limitations of traditional KDD techniques.

Our current research focuses on utilizing database semantics to reduce KDD processing time and to discover knowledge more robust to change in data.

For more information, please contact us.