Explore the vast range of titles available.

This paper presents and discusses various metrics proposed for evaluation of polyphonic sound event detection systems used in realistic situations where there are typically multiple sound sources active simultaneously. The system output in this case contains overlapping events, marked as multiple sounds detected as being active at the same time. The polyphonic system output requires a suitable procedure for evaluation against a reference. Metrics from neighboring fields such as speech recognition and speaker diarization can be used, but they need to be partially redefined to deal with the overlapping events. We present a review of the most common metrics in the field and the way they are adapted and interpreted in the polyphonic case. We discuss segment-based and event-based definitions of each metric and explain the consequences of instance-based and class-based averaging using a case study. In parallel, we provide a toolbox containing implementations of presented metrics.

In recent years, there has been a surge of interest in Artificial Intelligence (AI) systems that can provide human-centric explanations for decisions or predictions. No matter how good and efficient an AI model is, users or practitioners find it difficult to trust it if they cannot understand the AI model or its behaviours. Incorporating explainability that is human-centric in event detection systems is significant for building a decision-making process that is more trustworthy and sustainable. Human-centric and semantics-based explainable event detection will achieve trustworthiness, explainability, and reliability, which are currently lacking in AI systems. This paper provides a survey on human-centric explainable AI, explainable event detection, and semantics-based explainable event detection by answering some research questions that bother on the characteristics of human-centric explanations, the state of explainable AI, methods for human-centric explanations, the essence of human-centricity in explainable event detection, research efforts in explainable event solutions, and the benefits of integrating semantics into explainable event detection. The findings from the survey show the current state of human-centric explainability, the potential of integrating semantics into explainable AI, the open problems, and the future directions which can guide researchers in the explainable AI domain.

Sound Event Detection and Localization (SELD) is a comprehensive task that aims to solve the subtasks of Sound Event Detection (SED) and Sound Source Localization (SSL) simultaneously. The task of SELD lies in the need to solve both sound recognition and spatial localization problems, and different categories of sound events may overlap in time and space, making it more difficult for the model to distinguish between different events occurring at the same time and to locate the sound source. In this study, the Dual-conv Coordinate Attention Module (DCAM) combines dual convolutional blocks and Coordinate Attention, and based on this, the network architecture based on the two-stage strategy is improved to form the SELD-oriented Two-Stage Dual-conv Coordinate Attention Model (TDCAM) for SELD. TDCAM draws on the concepts of Visual Geometry Group (VGG) networks and Coordinate Attention to effectively capture critical local information by focusing on the coordinate space information of the feature map and dealing with the relationship between the feature map channels to enhance the feature selection capability of the model. To address the limitation of a single-layer Bi-directional Gated Recurrent Unit (Bi-GRU) in the two-stage network in terms of timing processing, we add to the structure of the two-layer Bi-GRU and introduce the data enhancement techniques of the frequency mask and time mask to improve the modeling and generalization ability of the model for timing features. Through experimental validation on the TAU Spatial Sound Events 2019 development dataset, our approach significantly improves the performance of SELD compared to the two-stage network baseline model. Furthermore, the effectiveness of DCAM and the two-layer Bi-GRU structure is confirmed by performing ablation experiments.

Predictive Maintenance (PdM) is one of the most important applications of advanced data science in Industry 4.0, aiming to facilitate manufacturing processes. To build PdM models, sufficient data, such as condition monitoring and maintenance data of the industrial application, are required. However, collecting maintenance data is complex and challenging as it requires human involvement and expertise. Due to time constraints, motivating workers to provide comprehensive labeled data is very challenging, and thus maintenance data are mostly incomplete or even completely missing. In addition to these aspects, a lot of condition monitoring data-sets exist, but only very few labeled small maintenance data-sets can be found. Hence, our proposed solution can provide additional labels and offer new research possibilities for these data-sets. To address this challenge, we introduce MEDEP, a novel maintenance event detection framework based on the Pruned Exact Linear Time (PELT) approach, promising a low false-positive (FP) rate and high accuracy results in general. MEDEP could help to automatically detect performed maintenance events from the deviations in the condition monitoring data. A heuristic method is proposed as an extension to the PELT approach consisting of the following two steps: (1) mean threshold for multivariate time series and (2) distribution threshold analysis based on the complexity-invariant metric. We validate and compare MEDEP on the Microsoft Azure Predictive Maintenance data-set and data from a real-world use case in the welding industry. The experimental outcomes of the proposed approach resulted in a superior performance with an FP rate of around 10% on average and high sensitivity and accuracy results.

Distributed acoustic sensing (DAS) in optical fibers detect dynamic strains or sound waves by measuring the phase or amplitude changes of the scattered light. This contrasts with other distributed (and more conventional) methods, such as distributed temperature (DTS) or strain (DSS), which measure quasi-static physical quantities, such as intensity spectrum of the scattered light. DAS is attracting considerable attention as it complements the conventional distributed measurements. To implement DAS in commercial applications, it is necessary to ensure a sufficiently high signal-noise ratio (SNR) for scattered light detection, suppress its deterioration along the sensing fiber, achieve lower noise floor for weak signals and, moreover, perform high-speed processing within milliseconds (or sometimes even less). In this paper, we present a new, real-time DAS, realized by using the time gated digital-optical frequency domain reflectometry (TGD-OFDR) method, in which the chirp pulse is divided into overlapping bands and assembled after digital decoding. The developed prototype NBX-S4000 generates a chirp signal with a pulse duration of 2 μs and uses a frequency sweep of 100 MHz at a repeating frequency of up to 5 kHz. It allows one to detect sound waves at an 80 km fiber distance range with spatial resolution better than a theoretically calculated value of 2.8 m in real time. The developed prototype was tested in the field in various applications, from earthquake detection and submarine cable sensing to oil and gas industry applications. All obtained results confirmed effectiveness of the method and performance, surpassing, in conventional SM fiber, other commercially available interrogators.

The reduction of sensor network traffic has become a scientific challenge. Different compression techniques are applied for this purpose, offering general solutions which try to minimize the loss of information. Here, a new proposal for traffic reduction by redefining the domains of the sensor data is presented. A configurable data reduction model is proposed focused on periodic duty–cycled sensor networks with events triggered by threshold. The loss of information produced by the model is analyzed in this paper in the context of event detection, an unusual approach leading to a set of specific metrics that enable the evaluation of the model in terms of traffic savings, precision, and recall. Different model configurations are tested with two experimental cases, whose input data are extracted from an extensive set of real data. In particular, two new versions of Send–on–Delta (SoD) and Predictive Sampling (PS) have been designed and implemented in the proposed data–domain reduction for threshold–based event detection (D2R-TED) model. The obtained results illustrate the potential usefulness of analyzing different model configurations to obtain a cost–benefit curve, in terms of traffic savings and quality of the response. Experiments show an average reduction of 76 % of network packages with an error of less than 1%. In addition, experiments show that the methods designed under the proposed D2R–TED model outperform the original event–triggered SoD and PS methods by 10 % and 16 % of the traffic savings, respectively. This model is useful to avoid network bottlenecks by applying the optimal configuration in each situation.

Sound Event Localization and Detection (SELD) is the task of spatial and temporal localization of various sound events and their classification. Commonly, multitask models are used to perform SELD. In this work, a deep learning network model named channel-wise ‘FusionNet’ is designed to perform the SELD task. The novel fusion layer is introduced into the regular Deep Neural Network (DNN), where the input is fed channel-wise, and the outputs of all channels are fused to form a new feature representation. The key contribution of this work is the neural network model which helps to retain the channel-wise information from the multichannel input along with the spatial and temporal information. The proposed network utilizes separable convolution blocks in the convolution layers, therefore, the complexity of the model is low in terms of both time and space. The features used as input are Mel-band energies for Sound Event Detection (SED) and intensity vectors for the Direction-of-Arrival (DOA) estimation. The proposed network’s fusion layer provides a better representation of features for both SED and DOA estimation tasks. Experiments are performed on the recordings of the First-order Ambisonic (FOA) array format of the TAU-NIGENS Spatial Sound Events 2020 dataset. An improved performance is achieved in terms of Error Rate (ER), DOA error, and Frame Recall (FR) has been observed in comparison to the state-of-the-art SELD systems.

Detecting anomalies in data is a vital task, with numerous high-impact applications in areas such as security, finance, health care, and law enforcement. While numerous techniques have been developed in past years for spotting outliers and anomalies in unstructured collections of multi-dimensional points, with graph data becoming ubiquitous, techniques for structured graph data have been of focus recently. As objects in graphs have long-range correlations, a suite of novel technology has been developed for anomaly detection in graph data. This survey aims to provide a general, comprehensive, and structured overview of the state-of-the-art methods for anomaly detection in data represented as graphs. As a key contribution, we give a general framework for the algorithms categorized under various settings: unsupervised versus (semi-)supervised approaches, for static versus dynamic graphs, for attributed versus plain graphs. We highlight the effectiveness, scalability, generality, and robustness aspects of the methods. What is more, we stress the importance of anomaly attribution and highlight the major techniques that facilitate digging out the root cause, or the ‘why’, of the detected anomalies for further analysis and sense-making. Finally, we present several real-world applications of graph-based anomaly detection in diverse domains, including financial, auction, computer traffic, and social networks. We conclude our survey with a discussion on open theoretical and practical challenges in the field.

In critical Internet of Things (IoT) application domains, such as the Defense Industry and Healthcare, false alerts have many negative effects, such as fear, disruption of emergency services, and waste of resources. Therefore, an alert must only be sent if triggered by a correct event. Nevertheless, IoT networks are exposed to intrusions, which affects event detection accuracy. In this paper, an Anomaly Detection System (ADS) is proposed in a smart hospital IoT system for detecting events of interest about patients’ health and environment and, at the same time, for network intrusions. Providing a single system for network infrastructure supervision and e-health monitoring has been shown to optimize resources and enforce the system reliability. Consequently, decisions regarding patients’ care and their environments’ adaptation are more accurate. The low latency is ensured, thanks to a deployment on the edge to allow for a processing close to data sources. The proposed ADS is implemented and evaluated while using Contiki Cooja simulator and the e-health event detection is based on a realistic data-set analysis. The results show a high detection accuracy for both e-health related events and IoT network intrusions.

Detecting eye movements in raw eye tracking data is a well-established research area by itself, as well as a common pre-processing step before any subsequent analysis. As in any field, however, progress and successful collaboration can only be achieved provided a shared understanding of the pursued goal. This is often formalised via defining metrics that express the quality of an approach to solving the posed problem. Both the big-picture intuition behind the evaluation strategies and seemingly small implementation details influence the resulting measures, making even studies with outwardly similar procedures essentially incomparable, impeding a common understanding. In this review, we systematically describe and analyse evaluation methods and measures employed in the eye movement event detection field to date. While recently developed evaluation strategies tend to quantify the detector’s mistakes at the level of whole eye movement events rather than individual gaze samples, they typically do not separate establishing correspondences between true and predicted events from the quantification of the discovered errors. In our analysis we separate these two steps where possible, enabling their almost arbitrary combinations in an evaluation pipeline. We also present the first large-scale empirical analysis of event matching strategies in the literature, examining these various combinations both in practice and theoretically. We examine the particular benefits and downsides of the evaluation methods, providing recommendations towards more intuitive and informative assessment. We implemented the evaluation strategies on which this work focuses in a single publicly available library: https://github.com/r-zemblys/EM-event-detection-evaluation .