ROC with Cost Pareto Frontier Feature Selection Using Search Methods

Cyber-physical systems (CPS) are finding increasing application in many domains. CPS are composed of sensors, actuators, a central decision-making unit, and a network connecting all of these components. The design of CPS involves the selection of these hardware and software components, and this design process could be limited by a cost constraint. This study assumes that the central decision-making unit is a binary classifier, and casts the design problem as a feature selection problem for the binary classifier where each feature has an associated cost. Receiver operating characteristic (ROC) curves are a useful tool for comparing and selecting binary classifiers; however, ROC curves only consider the misclassification cost of the classifier and ignore other costs such as the cost of the features. The authors previously proposed a method called ROC Convex Hull with Cost (ROCCHC) that is used to select ROC optimal classifiers when cost is a factor. ROCCHC extends the widely used ROC Convex Hull (ROCCH) method by combining it with the Pareto analysis for cost optimization. This paper proposes using the ROCCHC analysis as the evaluation function for feature selection search methods without requiring an exhaustive search over the feature space. This analysis is performed on 6 real-world data sets, including a diagnostic cyber-physical system for hydraulic actuators. The ROCCHC analysis is demonstrated using sequential forward and backward search. The results are compared with the ROCCH selection method and a popular Pareto selection method that uses classification accuracy and feature cost.

Scalable and Reliable Multi-dimensional Sensor Data Aggregation in Data Streaming Architectures

Ever-increasing amounts of data and requirements to process them in real time lead to more and more analytics platforms and software systems designed according to the concept of stream processing. A common area of application is processing continuous data streams from sensors, for example, IoT devices or performance monitoring tools. In addition to analyzing pure sensor data, analyses of data for entire groups of sensors often need to be performed. Therefore, data streams of the individual sensors have to be continuously aggregated to a data stream for a group. Motivated by a real-world application scenario of analyzing power consumption in Industry 4.0 environments, we propose that such a stream aggregation approach has to allow for aggregating sensors in hierarchical groups, support multiple such hierarchies in parallel, provide reconfiguration at runtime, and preserve the scalability and reliability qualities of stream processing techniques. We propose a stream processing architecture fulfilling these requirements, which can be integrated into existing big data architectures. As all state-of-the-art stream processing frameworks have to handle a trade-off between latency, resource-efficiency, and correctness, our proposed architecture can be configured for low latency and resource-efficient computation or for always ensuring correct results. To assist adopters in choosing appropriate configuration options, we provide an experimental comparison. We present a pilot implementation of our proposed architecture and show how it is used in industry. Furthermore, in experimental evaluations we show that our solution scales linearly with the amount of sensors and provides adequate reliability in the presence of faults.