Mining Latent Entity Structures from Massive Unstructured and Interconnected Data

The “big data ” era is characterized by an explosion of infor-mation in the form of digital data collections, ranging from scientific knowledge, to social media, news, and everyone’s daily life. Examples of such collections include scientific publications, enterprise logs, news articles, social media and general Web pages. Valuable knowledge about multi-typed entities is often hidden in the unstructured or loosely struc-tured but interconnected data. Mining latent structured in-formation around entities uncovers sematic structures from massive unstructured data and hence enables many high-impact applications. In this tutorial, we summarize the closely related liter-ature in database systems, data mining, Web, information extraction, information retrieval, and natural language pro-cessing, overview a spectrum of data-driven methods that extract and infer such latent structures, from an interdisci-plinary point of view, and demonstrate how these structures support entity discovery and management, data understand-ing, and some new database applications. We present three categories of studies: mining conceptual, topical and rela-tional structures. Moreover, we present case studies on real datasets, including research papers, news articles and social networks, and show how interesting and organized knowl-edge can be discovered by mining latent entity structures from these datasets