Abstract: Analysis from security bodies, research institutes, government organizations shows that the amount of data leak incidents is more now a day. Most of the data leak incidents is because of mistakes done by humans, such type of leaks can be classified as inadvertent data leak.In this paper we present a privacy preserving data-leak detection (DLD) method to solve the data leak issue such as human mistakes (data loss through e-mail). We presented the design method using fuzzy fingerprint (Message Digest5 algorithm) technique that enhances data privacy during data leak detection operation. This approach is based on fast and real-world one-way encryption on the sensitive data. The approach enables the data owner to safely delegate the detection operation to the DLD provider by generating set of message digests to reduce inadvertent data leak in network traffic. This method assists the organization or owner to know the details of the file whether the file is sensitive or not through displaying or sending an alert message.

Keywords: Data leak detection (DLD), Message Digest (MD5), Social security number (SSN), Electronic mail (E-Mail).