Enterprise Wisdom Captured Socially (Invited Paper) Charalampos Chelmis Department of Computer Science University of Southern California Los Angeles, CA chelmis@usc.edu Vikram Sorathia, Viktor K. Prasanna Ming Hsieh Department of Electrical Engineering University of Southern California Los Angeles, CA {vsorathi, prasanna}@usc.edu Abstract—Data availability in online social networks as well as the business world has lately not been an issue. Vast amounts of data are being generated by social networking users in the form of informal interactions. What has been an issue, is the transformation of data into useful information, that in time and with appropriate processing becomes knowledge. In this paper we examine knowledge generation under informal social com- munications, based on semantically enriched user-generated data and associated metadata. We dynamically capture users’ interests and expertise using such semantically enriched con- tent. Knowledge networks of users emerge, exhibiting collective intelligence. To capture such collective knowledge, we propose a novel knowledge base paradigm, which seamlessly integrates information from multiple platforms and facilitates knowledge extraction, mining, discovery and inferencing. Using semanti- cally enriched user profiles, we compute semantic similarity between users and content in a joint semantic space, driving numerous applications. I. I NTRODUCTION The Web is transforming. From a graph of static pages, it has rapidly grown into a medium in which users are creating, utilizing, distributing, and rating information. Online social networks and social media in particular have significantly revolutionized the way people are searching for information. Social browsing [1] has become a primary method by which users discover new content on the Web. The business world has not stayed unaffected by this phenomenal transformation. Microblogging capabilities have penetrated the enterprise environment [2] providing a medium for users to share day-to-day operational knowledge and domain knowledge, discuss about problem solving, relevant emerging techniques, applications and technolo- gies, trends, etc. Enterprise microblogging services mostly emphasize on the business perspective and therefore their content revolves around their main business and work cul- ture, work practices, and everyday problems (technical or otherwise related to business). Regardless of the latent incentives and random variables that drive social activity and populated content, online social networks and enterprise microblogging services share quite a few characteristics [3]. Structurally, both exhibit power- law, small-world and scale-free properties. Contextually, both demonstrate assortative mixing characteristics [4] with respect to lexical and topical alignment [5], [6]. Researchers have mostly focused on analyzing online social networks, even though social graphs have been mined out of numerous sources. E-mail traffic analysis [7], [8] has lead the extraction of unofficial social networks in enterprise context, in an effort to understand how information flow in the enterprise differs from online social networks. [9] argued that “information extracted from e-mails could prove useful in a knowledge management perspective”, as it would facilitate expert and community identification. Enterprises however do not solely rely on e-mail traffic to share information among coworkers. Microblogging services constantly gain ground, while more traditional media like SharePoint and Office Communicator are heavily utilized as part of question-answering and problem solving processes. Active Directory provides a formal structure for the bulk of e-mail and/or social traffic to follow. However informal communication through microblogging services span orga- nizational charts. [10] presented an API for gathering and sharing interpersonal connections across multiple services and demonstrated its potential value with a comprehensive qualitative analysis. Likewise, online social networks users do not solely rely on their social networks to share information with their friends. E-mails, chat services, and blogging, bookmarking and rating sites are some of the ways people share infor- mation with their social circles. Again, similarities between online social networks and enterprise microblogging capa- bilities become apparent. Corporate microblogging services, on the other hand, act as facilitators of knowledge search and integration. Exchanged messages may contain knowledge in the form of solutions to particular problems, or may provide links to multiple external sources, like textbooks, research papers, FAQs, and best practice documents. Knowledge in such cases may not have been formally represented initially. Only when a specific question is being asked and an expert answer is provided, knowledge can be formally modeled and cap- tured. However, informal communication is inherently noisy, both in terms of presentation (e.g. unstructured, ungrammat-