Search

Given a set D of objects, a reverse nearest neighbor (RNN) query returns the objects q in D such that o is closer to a query object q than to any other object in D, according to a certain similarity metric. The existing ANN solutions are not sufficient because they either 1) rely on precomputed information that is expensive to maintain in the presence of updates or 2) are applicable only when the data consists of ‘Euclidean objects’ and similarity is measured using the L2 norm. In this paper, we present the first algorithms for efficient RNN search in generic metric spaces. Our techniques require no detailed representations of objects, and can be applied as long as their mutual distances can be computed and the distance metric satisfies the triangle inequality. We confirm the effectiveness of the proposed methods with extensive experiments. [ABSTRACT FROM AUTHOR]

In a competitive market, business process improvement is a requirement for any organization. This improvement can only be achieved with the support of comprehensive systems that fully monitor business processes. We propose an occurrence—based approach to business process monitoring that provides a holistic perspective of system dynamics, lending support to evolution aspects. More specifically, we present a three-dimensional artifact, called Occurrence, in which structure, behavior, and guidance are considered simultaneously. Based on this, we define more complex structures, namely Occurrence Base and Occurrence Management System, which serve as scaffolding to develop business process monitoring systems. We also present a specific occurrence-based design strategy that, using an MDE approach, has been applied by our research group for the development of successful monitoring applications. [ABSTRACT FROM PUBLISHER]

Associative classification is a promising technique to build accurate classifiers. However, in large or correlated data sets, association rule mining may yield huge rule sets. Hence, several pruning techniques have been proposed to select a small subset of high-quality rules. Since the availability of a ‘rich’ rule set may improve the accuracy of the classifier, we argue that rule pruning should be reduced to a minimum. The L3 associative classifier is built by means of a lazy pruning technique that discards exclusively rules that only misclassify training data. The classification of unlabeled data is performed in two steps. A small subset of high-quality rules is first considered. When this set is not able to classify the data, a larger rule set is exploited. This second set includes rules usually discarded by previous approaches. To cope with the need of mining large rule sets and to efficiently use them for classification, a compact form is proposed to represent a complete rule set in a space-efficient way and without information loss. An extensive experimental evaluation on real and synthetic data sets shows that L3 improves the classification accuracy with respect to previous approaches; [ABSTRACT FROM AUTHOR]

In reality, spatial objects (e.g., hotels) not only have spatial locations but also have quality attributes (e.g., price, star). An object p is said to dominate another one p^\prime , if p is no worse than p^\prime with respect to every quality attribute and p is better on at least one quality attribute. Traditional spatial queries (e.g., nearest neighbor, closest pair) ignore quality attributes, whereas conventional dominance-based queries (e.g., skyline) neglect spatial locations. Motivated by these observations, we propose a novel query by combining spatial and quality attributes together meaningfully. Given a set of (competitors') spatial objects P, a set of (candidate) locations L, and a quality vector \Psi as design competence (for L), the farthest dominated location (FDL) query retrieves the location s \in L such that the distance to its nearest dominating object in P is maximized. FDL queries are suitable for various spatial decision support applications such as business planning, wild animal protection, and digital battle field systems. As FDL queries cannot be readily solved by existing techniques, we develop several efficient R-tree-based algorithms for processing FDL queries, which offer users a range of selections in terms of different indexes available on the data. We also generalize our methods to support the generic distance metric and other interesting query types. The experimental results on both real and synthetic data sets disclose the performance of those algorithms, and reveal the most efficient and scalable one among them. [ABSTRACT FROM PUBLISHER]