Abstract:

If the database is updated frequently, Conditional Functional Dependencies (CFDs) that have met the conditions may changes.. In order to obtain accurate CFDs, we can rerun the discovering process over the updated database. However, it spends a lot of time on dealing with the original dataset. Aiming at this problem, based on CFINDER algorithm, the paper proposed an incremental discovering algorithm for CFDs, which is named as CFUP. When a batch of new data is added to the database, the CFUP algorithm scans dataset to decide whether existing CFDs is valid or not, and the new data produces new CFDs, to achieve an incremental update for CFDs. Experiments show that the CFUP algorithm can effectively find CFDs by using information from the last discovering process. Compared with the rerun of the CFINDER algorithm, it can reduce the scanning number to original dataset and improve the efficiency of discovering CFDs.

Key words: conditional functional dependencies;incremental algorithm;database