Guest Editorial: Special Section on the International Conference on Data Engineering

THE 28th International Conference on Data Engineering was held in Washington, DC, on April 1-5, 2012. ICDE 2012 attracted 413 submissions in the research track, 19 submissions in the, industrial track, and 68 submissions in the demo track. After a thorough review from the corresponding committees, 100 research papers, six industrial contributions, and 28 demo proposals were selected for inclusion in the conference program. This special section consists of journal versions of 10 outstanding papers selected among the 100 accepted research contributions. All papers were revised and substantially extended, over their conference versions and went through a rigorous review process to ensure the high quality standards of the IEEE Transactions on Knowledge and Data Engineering. They cover a broad range of topics highlighting the liveliness of the data engineering field. The paper “Answering Why-Not Questions on Top-K Queries” by Zhian He and Eric Lo extends the problem, of explaining missing tuples in query results, known as the why-not problem, for the case of top-k queries. The authors develop algorithms for processing such why-not questions efficiently and present case studies and experimental results to demonstrate the efficiency of their approach in providing high quality explanations. The paper “BestPeerþþ: A Peer-to-Peer Based LargeScale Data Processing Platform” by Gang Chen, Tianlei Hu, Dawei Jiang, Peng Lu, Kian-Lee Tan, Hoang Tam Vo, and Sai Wu presents a peer-to-peer data management system that provides elastic data sharing services, for corporate data processing applications in the cloud. The system was evaluated on the Amazon EC2 platform and shown to outperform HadoopDB on enterprise workloads. The paper “Discovering Conservation Rules” by Lukasz Golab, Howard Karloff, Flip Korn, Barna Saha, and Divesh Srivastava proposes a new class of constraints, termed Conversation Rules (CRs), for data quality analysis in monitoring infrastructure networks. The authors provide confidence metrics to quantify the strength of a CR and approximation algorithms for the problem of discovering a concise summary of the subset of the data that satisfies a given CR. The utility of CRs is demonstrated through a set of experiment with real data. The paper “A Generic Framework for Top-k Pairs and Top-k Objects Queries over Sliding Windows” by Zhitao Shen, Muhammad Aamir Cheema, Xuemin Lin, Wenjie Zhang, and Haixun Wang provides a general framework for processing multiple top-k object and top-k pair queries over sliding windows. Given a function that computes a score between two objects, a top-k pair query returns the k pairs with the largest score. A detailed complexity analysis is presented along with experimental results. The paper “Incremental Detection of Inconsistencies in Distributed Data” by Wenfei Fan, Jianzhong Li, Nan Tang, and Wenyuan Yu focuses on incremental detection of functional dependency violations in distributed data. Given a distributed database partitioned either vertically or horizontally, a set of conditional functional dependencies (CFDs), a set of violations of the CFDs, and updates of the database, the problem studied is locating with minimum data shipment, changes to the set of violations in response to changes in the database. They show that the incremental detection problem is NP-complete and provide appropriate algorithms. The paper “LARS : An Efficient and Scalable LocationAware Recommender System” by Mohamed Sarwat, Justin J. Levandoski, Ahmed Eldawy, and Mohamed F. Mokbel presents a recommender system that incorporates the notion of location. It combines spatial ratings for non-spatial objects, non-spatial ratings for spatial objects, and special ratings for spatial objects. The authors also exploit user rating locations to influence recommendations with ratings spatially close to the query. Experimental results are presented using real data from Foursquare and MovieLens. The paper “Mapping XML to a Wide Sparse Table” by Liang Jeff Chen, Philip A. Bernstein, Peter Carlin, Dimitrije Filipovic, Michael Rys, Nikita Shamgunov, James F. Terwilliger, Milos Todic, Sasa Tomasevic, and Dragan Tomic proposes a novel mapping of XML data into one wide table with sparsely populated columns that provides good performance for workloads observed in enterprise applications but not supported efficiently by existing work. Rewrite rules that minimize the number of joins are also presented. The paper “On the Use of Side Information for Mining Text Data” by Charu C. Aggarwal, Yuchen Zhao, and Philip S. Yu presents a clustering algorithm for documents which combines classical partitioning algorithms with probabilistic models and then they show how to extend the approach to the classification problem. Experimental results on a number of real data sets are presented. The paper “PrefDB: Supporting Preferences as FirstClass Citizens in Relational Databases” by Anastasios Arvanitis and Georgia Koutrika makes the case for placing preference aware query processing closer to the DBMS. The authors propose a preference-aware relational data model and an associated algebra. They also present query J. Gehrke is with Cornell University and Microsoft. B.C. Ooi is with the National University of Singapore, Singapore. E. Pitoura is with the University of Ioannina, Greece.