This is paper highlights the problem of digital identity, or cross-set unique identifying tokens, inherent, ill the application of social software hi business processes. As social software, via blogs, wikis and other Web 2.0 software, starts to bring value to the enterprise, there is a need for a unified digital identity resource. From this basis, concepts of reputation and trust may be leveraged ill the Context Of human agents working with such software and in larger workflow processes. By analysing human agent activity within existing data sets and providing a mechanism for adding new data, it is possible to correlate human agent activities and data, creation via a digital identity across disparate sources of data. From this basis, new business processes may be created using this deeper understanding of human agents. This paper highlights the need for digital identity and presents a novel method for extracting digital identity Pattern,, from heterogeneous data. sets ill automated manner.