Abstract
When an attacker or a malicious user updates a database, the resulting damage can spread to other parts of the database through valid users. A fast and accurate damage assessment must be performed as soon as such an attack is detected. In this paper, we have discussed two approaches for damage assessment in an affected dataabbase. While the first one uses transaction dependency relationships to determine affected transactions, the second approach considers data dependency relationships to identify affected data items for future recovery. These relationships are stored in a matrix format for faster manipulation.