Special Section on the International Conference on Data Engineering 2015

THIS special section covers extended versions of topranked papers of the 31st International Conference on Data Engineering that was held in Seoul, Korea, on April 1317, 2015. The ICDE 2015 conference received 645 submissions within the research track. With an acceptance rate of 13 percent—which reflects the goal of ICDE to keep the highest quality standards—84 research papers were accepted. Within the extremely well structured reviewing process, every paper was at least reviewed by three members of the program committee (PC). The PC consisted of 142 members; this year, special emphasis was laid on a balanced structure of the committee by appointing experienced senior researchers as well as young researchers starting out in a career in academia. Every PC member’s review workload was about eight and nine papers. In addition, every paper was assigned to a topic area. ICDE 2015 was structured into 15 topic areas handled by individual area chairs. With topics ranging from Big Data system architectures via modern hardware support to temporal, spatial mobile data management, ICDE 2015 provided a rich bouquet of specialized areas managed by internationally renowned experts of the field. In addition to research papers, the conference also offered an industrial track with 15 papers accepted out of 33 submissions, an application track with four papers (11 submissions) as well as a demo track with 73 submissions and 29 accepted demos resulting in an acceptance rate of 39.7 percent. Out of the 84 papers presented at the conference, five outstanding papers were selected and the authors were offered the opportunity to provide an extended version to be published in this special section of the IEEE Transactions on Knowledge and Data Engineering. After an additional—and again—rigorous review process, all papers are now substantially extended in comparison to their conference version. Those five papers cover a wide range of topics and very nicely reflect the breadth of research work in the context of datamanagement. The paper “Understand Short Texts by Harvesting and Analyzing Semantic Knowledge” by Wen Hua, Zhongyuan Wang, Haixun Wang, Kai Zheng, and Xiaofang Zhou addresses the challenge of how to understand short texts, which is crucial to many applications, because they do not always observe the syntax of a written language or do not contain sufficient statistical signals to support many stateof-the-art approaches for text mining such as topic modeling. The authors present a prototype system for short text understanding which exploits semantic knowledge provided by a well-known knowledgebase automatically harvested from a web corpus including a comprehensive performance evaluation on real-life data. The paper “NetworkMotif Discovery: A GPUApproach” by Wenqing Lin, Xiaokui Xiao, Xing Xie, and Xiao-Li Li tackles the challenge of identifying networkmotifs as important applications in numerous domains. This paper presents a novel study on network motif discovery using GPUs in order to parallelize a large number of subgraph matching tasks in computing subgraph frequencies from random graphs. The paper explores the design space of GPU-based subgraph matching algorithms, with careful analysis of several crucial factors (such as branch divergences and memory coalescing) that affect the performance of GPU programs. The paper “Bump Hunting in the Dark: Local Discrepancy Maximization on Graphs” by Aristides Gionis, Michael Mathioudakis, and Antti Ukkonen studies the problem of discrepancy maximization on graphs to identify a connected subgraph that containsmanymore nodes from a set of nodes of an underlying graph than other nodes. Considering two different access models, the authors prove that the basic problem of discrepancy maximization on graphs is NP-hard and empirically evaluate the performance of four heuristics for solving it. The paper “Enumeration of Maximal Cliques from an Uncertain Graph” by ArkoMukherjee, Pan Xu, and Srikanth Tirthapura considers the enumeration of dense substructures (maximal cliques) from an uncertain graph. The paper presents matching upper and lower bounds on the number of a-maximal cliques possible within an uncertain graph by providing an algorithm to enumerate a-maximal cliques whose worst-case runtime is near-optimal. An experimental evaluation shows the practical utility of the algorithm. The paper “Efficient Metric Indexing for Similarity Search and Similarity Joins” by Lu Chen, Yunjun Gao, Xinhan Li, Christian S. Jensen, and Gang Chen proposes a versatile and efficient disk-based index for metric data, the space-filling curve, and pivot-based Bþ-tree called SPB-tree in order to better support spatial queries including similarity search and similarity joins. The paper covers the W. Lehner is with the Department of Computer Science, Technische Universit€ at Dresden, Dresden D-01062, Germany. E-mail: [email protected]. J. Gehrke is with Microsoft, Destiny, NY 13204. E-mail: [email protected]. K. Shim is with the Department of Electrical and Computer Engineering, Seoul National University, Seoul 151-742, Korea. E-mail: [email protected].