Search

Complex event processing (CEP) evaluates queries over streams of event data to detect situations of interest. If the event data are produced by geographically distributed sources, CEP may exploit in-network processing that distributes the evaluation of a query among the nodes of a network. To this end, a query is modularized and individual query operators are assigned to nodes, especially those that act as data sources. Existing solutions for such operator placement, however, are limited in that they assume all query results to be gathered at one designated node, commonly referred to as a sink. Hence, existing techniques postulate a hierarchical structure of the network that generates and processes the event data. This largely neglects the optimisation potential that stems from truly decentralised query evaluation with potentially many sinks. To address this gap, in this paper, we propose Multi-Sink Evaluation (MuSE) graphs as a formal computational model to evaluate common CEP queries in a decentralised manner. We further prove the completeness of query evaluation under this model. Striving for distributed CEP that can scale to large volumes of high-frequency event streams, we show how to reason on the network costs induced by distributed query evaluation and prune inefficient query execution plans. As such, our work lays the foundation for distributed CEP that is both, sound and efficient. [ABSTRACT FROM AUTHOR]

Process mining provides a rich set of techniques to discover valuable knowledge of business processes based on data that was recorded in different types of information systems. It enables analysis of end‐to‐end processes to facilitate process re‐engineering and process improvement. Process mining techniques rely on the availability of data in the form of event logs. In order to enable process mining in diverse environments, the recorded data need to be located and transformed to event logs. The journey from raw data to event logs suitable for process mining can be addressed by a variety of methods and techniques, which are the focus of this article. In particular, techniques proposed in the literature to support the creation of event logs from raw data are reviewed and classified. This includes techniques for identification and extraction of the required event data from diverse sources as well as their correlation and abstraction. This article is categorized under:Technologies > Structure Discovery and ClusteringFundamental Concepts of Data and Knowledge > Data ConceptsTechnologies > Data Preprocessing [ABSTRACT FROM AUTHOR]

Urban mobility impacts urban life to a great extent. To enhance urban mobility, much research was invested in traveling time prediction: given an origin and destination, provide a passenger with an accurate estimation of how long a journey lasts. In this work, we investigate a novel combination of methods from Queueing Theory and Machine Learning in the prediction process. We propose a prediction engine that, given a scheduled bus journey (route) and a ‘source/destination’ pair, provides an estimate for the traveling time, while considering both historical data and real-time streams of information that are transmitted by buses. We propose a model that uses natural segmentation of the data according to bus stops and a set of predictors, some use learning while others are learning-free, to compute traveling time. Our empirical evaluation, using bus data that comes from the bus network in the city of Dublin, demonstrates that the snapshot principle, taken from Queueing Theory, works well yet suffers from outliers. To overcome the outliers problem, we use Machine Learning techniques as a regulator that assists in identifying outliers and propose prediction based on historical data. [ABSTRACT FROM AUTHOR]

Information systems support the execution of business processes. As part of that, data about process execution is recorded in event logs, which can be used to analyse the control-flow of the respective processes. However, such data may contain personal information on process stakeholders that is protected by privacy regulations. Process analysis based on event logs shall, therefore, employ anonymization techniques. In this paper, we introduce two approaches to anonymize the recorded control-flow of a process. Specifically, we present SaCoFa and SaPa as two techniques to anonymize the result of trace-variant queries over an event log. Unlike existing techniques that achieve differential privacy through randomized noise insertion, our techniques rely on noise insertion mechanisms that incorporate a process' semantics, thereby avoiding easily-recognizable noise. Both techniques take different design choices, though. SaCoFa anonymizes a trace-variant distribution directly, thereby focusing on utility preservation at the expense of potentially changing the number of a traces in the result considerably. SaPa, in turn, anonymizes a trace-variant distribution indirectly, through play-out of an anonymized directly-follows distribution. This way, the number of traces in the result is close to the original log, but the drop in utility may become larger due to using only local control-flow information. However, our experiments demonstrate that both approaches strike a better balance of preserving the utility of an event log compared to existing techniques. • We propose methods to anonymize the control-flow perspective of business processes. • The anonymized data is protected by differential privacy. • Our techniques provide better utility for process discovery than the state of the art. [ABSTRACT FROM AUTHOR]