Results 1 
3 of
3
Anomaly detection in dynamic networks: a survey
 Wiley Interdisciplinary Reviews: Computational Statistics
, 2015
"... Anomaly detection is an important problem with multiple applications, and thus has been studied for decades in various research domains. In the past decade there has been a growing interest in anomaly detection in data represented as networks, or graphs, largely because of their robust expressivene ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Anomaly detection is an important problem with multiple applications, and thus has been studied for decades in various research domains. In the past decade there has been a growing interest in anomaly detection in data represented as networks, or graphs, largely because of their robust expressiveness and their natural ability to represent complex relationships. Originally, techniques focused on anomaly detection in static graphs, which do not change and are capable of representing only a single snapshot of data. As realworld networks are constantly changing, there has been a shift in focus to dynamic graphs, which evolve over time. In this survey, we aim to provide a comprehensive overview of anomaly detection in dynamic networks, concentrating on the stateoftheart methods. We first describe four types of anomalies that arise in dynamic networks, providing an intuitive explanation, applications, and a concrete example for each. Having established an idea for what constitutes an anomaly, a general twostage approach to anomaly detection in dynamic networks that is common among the methods is presented. We then construct a twotiered taxonomy, first partitioning the methods based on the intuition behind their approach, and subsequently subdividing them based on the types of anomalies they detect. Within each of the tier one categoriescommunity, compression, decomposition, distance, and probabilistic model basedwe highlight the major similarities and differences, showing the wealth of techniques derived from similar conceptual approaches. © 2015 The Authors. financial systems connecting banks across the world, electric power grids connecting geographically distributed areas, and social networks that connect users, businesses, or customers using relationships such as friendship, collaboration, or transactional interactions. These are examples of dynamic networks, which, unlike static networks, are constantly undergoing changes to their structure or attributes. Possible changes include insertion and deletion of vertices (objects), insertion and deletion of edges (relationships), and modification of attributes (e.g., vertex or edge labels). WIREs Computational Statistics An important problem over dynamic networks is anomaly detectionfinding objects, relationships, or
IOS Press The Role of Semantics in Smart Cities Producing Linked Data for Smart Cities: the
"... Semantic Web technologies and in particular Linked Open Data provide a means for sharing knowledge about cities as physical, social, and technical systems, so enabling the development of smart city applications. This paper presents the case of Catania with the aim of sharing the lessons learnt, whi ..."
Abstract
 Add to MetaCart
(Show Context)
Semantic Web technologies and in particular Linked Open Data provide a means for sharing knowledge about cities as physical, social, and technical systems, so enabling the development of smart city applications. This paper presents the case of Catania with the aim of sharing the lessons learnt, which can be reused as reference practices in other cases with similar requirements. The importance of achieving syntactic as well as semantic interoperability as a result of transforming heterogeneous sources into Linked Data is discussed: semantic interoperability must be solved at data level in order to ease the development of smart city applications. This claim is supported by showing how this issue impacts on the design of two smart city applications. As main contributions, the paper describes: (i) methods, procedures, and tools used for transforming heterogeneous sources into Linked Data; (ii) an ontology design pattern for modelling urban public transportation routes; (iii) methods, procedures and tools for ensuring semantic interoperability during the transformation process; (iv) the design of two smart city applications based on Linked Data. All produced data, models, and prototypes are publicly accessible online.
Mining evolving network processes Supplemental material
"... The singlesnapshot MINESMOOTH problem is called Heaviest Subgraph (HS) and has been shown to be equivalent to the Prize Collecting Steiner Tree (PCST) problem1 [5], which considers nodes with nonnegative weights (prizes) and edges with negative weights (costs). An HS instance can be reduced to ..."
Abstract
 Add to MetaCart
(Show Context)
The singlesnapshot MINESMOOTH problem is called Heaviest Subgraph (HS) and has been shown to be equivalent to the Prize Collecting Steiner Tree (PCST) problem1 [5], which considers nodes with nonnegative weights (prizes) and edges with negative weights (costs). An HS instance can be reduced to a PCST instance by (i) considering only nonnegative edges and computing connected components in the resulting graph, and then (ii) collapsing each nonnegative component into a single node that accumulates all corresponding positive edge weights [2]. Note that in PCST the node prize can be zero (a zero node) or positive (a positive node). The rooted PCST problem further imposes that a particular node (root) belongs to the solution. Both PCST and rooted PCST are NPhard and the latter cannot be approximated within any constant factor [4]. However, both ones can be solved in linear time on trees [6]. II. NPHARDNESS OF αMINESMOOTH