Data Leakage Prevention by Using Word Gram Based Classification and Clustering

Abstract: Nowadays keeping one’s data safe and secure has become very difficult task. The theft and misuse of the data has become a reason of concern ranging from small kids to adults, from small scale industries to multi-national industries. So to prevent this data leakage is extremely necessary. There were many techniques invented like firewalls, anti-viruses, watermarking etc. to prevent this leakage but eventually all tend to fail. Furthermore another techniques were invented known as the Data Leakage Prevention Software (DLPs) which keep the track of data while they are being transferred in the network. We have done research on one such technique known as the word gram. In this paper, the different uses and methods of using the word gram has been shown. This technique classifies the data packets before sending them in the network and makes clusters so while receiving the clusters can be compared with the sent clusters and we can be sure that the data hasn’t been altered or interfered with during the transfer.

Keywords: Word Gram, Data Leakage and Prevention, Clustering, Frequency Matching, N-gram