Guest Editors' Introduction: Special Section on Intelligence and Security Informatics

THE past several years have witnessed significant interest in security-related research in a wide range of application context spanning across homeland security, national and international security, economic and societal security, to personal and community security. A number of Information Technology-related academic disciplines including but not limited to information and computer sciences, information systems, human-computer studies, technology adoption, and policy studies have been making rapid progress in developing and evaluating customized frameworks, methodologies, techniques, and systems to meet specific information processing and knowledge management challenges arisen in security-related applications. An emerging field of cross-disciplinary study, Intelligence and Security Informatics (ISI), encompasses these efforts through an integrated technological, organizational, and policy-based approach. The ISI research community is rapidly maturing. The IEEE has been sponsoring the flagship ISI annual international conference series, which started in 2003. Technical workshops focusing on ISI topics are being held regularly in Pacific Asia and Europe. Please visit http:// www.isiconference.org/ for a list of ISI conferences and workshops. Most of the past ISI conference and workshop proceedings have been published in the Springer Lecture Notes in Computer Science series. In 2007 and 2008, the IEEE Press published the Proceedings of the IEEE ISI Conference. As the body of ISI literature continues to grow, we see a critical need to publish a high-quality collection of academic works on various ISI topics to provide an integrated and synthesized view of the current state of the art, identify challenges and opportunities for future work, and further promote community-building among researchers with previously disparate backgrounds and reference disciplines. This IEEE Transactions on Knowledge and Data Engineering special section on ISI serves this critical need with an emphasis on work employing research methodologies from the Knowledge and Data Engineering community. In response to the special section call for papers, 44 papers were submitted. Among these submissions, seven regular papers and five concise papers were accepted for publication. With a few exceptions, most of these papers have gone through two rounds of reviews and revisions; however, several papers did go through a third round of review. As special section editors, we are very impressed by the technical quality and application relevance of these papers and appreciate the significant efforts of the authors and reviewers to make this special section a high-quality snapshot of the state of the art ISI research. Based on technical topics covered, these 12 ISI papers can be roughly classified into the following three groups: information infrastructure and data security, adversarial data and text mining, and innovative applications and decision-making. There are three contributions in the “information infrastructure and data security” group. The paper titled “Protection of Database Security via Collaborative Inference Detection,” by Yu Chen and Wesley W. Chu, proposes a database security mechanism that will prevent single users or user groups from inferring sensitive information from a series of seemingly innocuous database queries. At the core of their mechanism is a probabilistic semantic inference model that captures all possible inference channels from any data attribute to sensitive attributes that need to be protected. The authors propose an efficient computational mechanism to derive the semantic inference model and study through computational experiments various factors related to collaborative inference by user groups and the detection of such collaborative activities. The second paper by Nan Zhang and Wei Zhao, “Privacy Protection against Malicious Adversaries in Distributed Information Sharing Systems,” aims to address privacy protection challenges in distributed information sharing systems without a trusted third-party mechanism. The application context of this research involves distributed settings in which multiple autonomous entities are willing to share certain information without disclosing their private data. The authors consider two classes of adversarial entities in this information sharing game: weakly malicious adversaries and strongly malicious adversaries, and design corresponding privacypreserving protocols. Formal analyses concerning various properties of these protocols are presented. The third and IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 20, NO. 8, AUGUST 2008 1009