There are many modes of communication, but electronic communication is the most noted one in present. Internet is the backbone for all these communications. In digital forensics, finding out the author of a document is a big qestion, identity of the author, their demographic background, and how they are linked to other documents. So major challenges in digital forensic investigation are author identification of message(s) and non-repudiation. In this paper we used Stylometry based human writing feature extraction as a solution for the author identification problem. Stylometry is not only a way of human writing pattern identification but it can also be used for human gender identification. This paper is oriented to highlight some of the ways to manage such problems like anonymous email messages, email abuse and even for the digital forensics. In this paper, 62 stylistic features have been collected for different users, using C language. 22 samples of 150 words for each user have been taken to train the Neural Network using Back Propagation Algorithm(BPA). In different variations of the experimental setup, 98.312% accuracy have been achieved.