Search

A finite point process approach to multi-target localization from a transient signal is presented. After modeling the measurements as a Poisson point process, we propose a twofold scheme that includes an expectation maximization algorithm to estimate the target locations for a given number of targets and an information theoretic algorithm to select the number of targets. The proposed localization scheme does not require explicitly solving the data association problem and can account for clutter noise as well as missed detections. Although point process theory has been widely utilized for sequential tracking of multiple moving targets, the application of point process theory for multi-target localization from transient measurements has received very little attention. The optimal subpattern assignment metric is used to assess the performance and accuracy of the proposed localization algorithm. Implementation of the proposed algorithm on synthetic data yields desirable results. The proposed algorithm is then applied to the multi-shooter localization problem using acoustic gunfire detection systems. [ABSTRACT FROM AUTHOR]

To enable reliable crowdsourcing applications, it is of great importance to develop algorithms that can automatically discover the truths from possibly noisy and conflicting claims provided by various information sources. In order to handle crowdsourcing applications involving big or streaming data, a desirable truth discovery algorithm should not only be effective, but also be scalable. However, with respect to quantitative crowdsourcing applications such as object counting and percentage annotation, existing truth discovery algorithms are not simultaneously effective and scalable. They either address truth discovery in categorical crowdsourcing or perform batch processing that does not scale. In this paper, we propose new parallel and streaming truth discovery algorithms for quantitative crowdsourcing applications. Through extensive experiments on real-world and synthetic datasets, we demonstrate that 1) both of them are quite effective, 2) the parallel algorithm can efficiently perform truth discovery on large datasets, and 3) the streaming algorithm processes data incrementally, and it can efficiently perform truth discovery both on large datasets and in data streams. [ABSTRACT FROM AUTHOR]

Truth discovery is an important technique for enabling reliable crowdsourcing applications. It aims to automatically discover the truths from possibly conflicting crowdsourced claims. Most existing truth discovery approaches focus on categorical applications, such as image classification. They use the accuracy, i.e., rate of exactly correct claims, to capture the reliability of participants. As a consequence, they are not effective for truth discovery in quantitative applications, such as percentage annotation and object counting, where similarity rather than exact matching between crowdsourced claims and latent truths should be considered. In this paper, we propose two unsupervised Quantitative Truth Finders (QTFs) for truth discovery in quantitative crowdsourcing applications. One QTF explores an additive model and the other explores a multiplicative model to capture different relationships between crowdsourced claims and latent truths in different classes of quantitative tasks. These QTFs naturally incorporate the similarity between variables. Moreover, they use the bias and the confidence instead of the accuracy to capture participants’ abilities in quantity estimation. These QTFs are thus capable of accurately discovering quantitative truths in particular domains. Through extensive experiments, we demonstrate that these QTFs outperform other state-of-the-art approaches for truth discovery in quantitative crowdsourcing applications and they are also quite efficient. [ABSTRACT FROM PUBLISHER]

Subjective Logic (SL) is a type of probabilistic logic, which is suitable for reasoning about situations with uncertainty and incomplete knowledge. In recent years, SL has drawn a significant amount of attention from the multi-agent systems community as it connects beliefs and uncertainty in propositions to a rigorous statistical characterization via Dirichlet distributions. However, one serious limitation of SL is that the belief updates are done only based on completely observable evidence. This work extends SL to incorporate belief updates from partially observable evidence. Normally, the belief updates in SL presume that the current evidence for a proposition points to only one of its mutually exclusive attribute states. Instead, this work considers that the current attribute state may not be completely observable, and instead, one is only able to obtain a measurement that is statistically related to this state. In other words, the SL belief is updated based upon the likelihood that one of the attributes was observed. The paper then illustrates properties of the partial observable updates as a function of the state likelihood and illustrates the use of these likelihoods for a trust estimation application. Finally, the utility of the partial observable updates is demonstrated via various simulations including the trust estimation case. [ABSTRACT FROM AUTHOR]

An investigation of the feasibility of a mesh network of proximity sensors to track targets is presented. In such a network the sensors report binary detection/nondetection measurements for the targets within proximity. A new probability hypothesis density (PHD) filter and its particle implementation for multiple-target tracking in a proximity sensor network are proposed. The performance and robustness of the new method are evaluated over simulated matching and mismatching cases for the sensor models. The simulations demonstrate the utility of the PHD filter to both track the number of targets and their locations. [ABSTRACT FROM AUTHOR]

An optimal strategy for geometric sensor placement to enhance target tracking performance is developed. Recently, a considerable amount of work has been published on optimal conditions for single-update placement of homogeneous sensors (same type and same measurement quality) in which the targets are either assumed perfectly known or the target location uncertainty is averaged out via the expected value of the determinant of the Fisher information matrix (FIM). We derive conditions for optimal placement of heterogeneous sensors based on maximization of the information matrix to be updated by the heterogeneous sensors from an arbitrary Gaussian prior characterizing the uncertainty about the initial target location. The heterogeneous sensors can be of the same or different types (ranging sensors, bearing-only sensors, or both). The sensors can also make, over several time steps, multiple independent measurements of different qualities. Placement strategies are derived and their performance is illustrated via simulation examples. [ABSTRACT FROM AUTHOR]

This paper introduces a novel method to score how well proposed fused image quality measures (FIQMs) indicate the effectiveness of humans to detect targets in fused imagery. The human detection performance is measured via human perception experiments. A good FIQM should relate to perception results in a monotonic fashion. The method computes a new diffuse prior monotonic likelihood ratio (DPMLR) to facilitate the comparison of the H_1 hypothesis that the intrinsic human detection performance is related to the FIQM via a monotonic function against the null hypothesis that the detection and image quality relationship is random. The paper discusses many interesting properties of the DPMLR and demonstrates the effectiveness of the DPMLR test via Monte Carlo simulations. Finally, the DPMLR is used to score FIQMs with test cases considering over 35 scenes and various image fusion algorithms. [ABSTRACT FROM AUTHOR]

Abstract: This paper develops a block Wiener-based image registration method to align consecutive frames in a video sequence. The main purpose of this registration technique is to serve as a processing step in a moving target indication (MTI) algorithm, where subsequent steps include frame-to-frame differencing and thresholding. Experiments show that the block Wiener-based approach is computationally efficient and effective in terms of facilitating a MTI. These experiments compare the block Wiener-based against traditional global parametric registration, pixel-based Wiener registration, block-based motion compensation, and state-of-the-art phase-based registration. [Copyright &y& Elsevier]

Successful track-to-track data association in a multisensor, multitarget scenario is predicated on a proper cost function. The cost function for associating two tracks is well established. This paper investigates different versions of the likelihood that more than two tracks represent the same target. These likelihoods lead to different cost functions for the general track-to-track association problem. Two of the likelihoods are approximations based on ad-hoc extensions of the well-known two-track expression. The next approximation is the generalized likelihood (GL), where the true state is replaced by the maximum likelihood (ML) estimate of the target state in the true likelihood. The final likelihood is derived exactly from a diffuse prior of the target state and is referred to as the diffuse prior likelihood (DPL). This paper reveals the connection between the DPL and GL. The last two likelihood expressions incorporate the cross-track error covariance matrices, which are not readily available in distributed track fusion. Therefore, the paper considers three forms of the DPL or CL: 1) the correlated tracks (CT) form by using the actual cross-covariance matrices, 2) the independent tracks (IT) form by assuming the cross-covariance matrices are zero, and 3) the approximated correlated tracks (ACT) form by approximating the cross-covariance matrices. Simulations compare the performance of all likelihood versions. [ABSTRACT FROM AUTHOR]

This paper introduces a new direction-of-arrival (DOA) estimation algorithm for wideband sources called test of orthogonality of projected subspaces (TOPS). This new technique estimates DOAs by measuring the orthogonal relation between the signal and the noise subspaces of multiple frequency components of the sources. TOPS can be used with arbitrary shaped one-dimensional (1-D) or two-dimensional (2-fl) arrays. Unlike other coherent wideband methods, such as the coherent signal subspace method (CSSM) and WAVES, the new method does not require any preprocessing for initial values. The performance of those wideband techniques and incoherent MUSIC is compared with that of the new method through computer simulations. The simulations show that this new technique performs better than others in mid signal-to-noise ratio (SNR) ranges, while coherent methods work best in low SNR and incoherent methods work best in high SNR. Thus, TOPS fills a gap between coherent and incoherent methods. [ABSTRACT FROM AUTHOR]

This work considers the problem of selecting the best nodes for localizing (in the mean squared (MS) position error sense) a target in a distributed wireless sensor network. Each node consists of an array of sensors that are able to estimate the direction of arrival (DOA) to a target. Different computationally efficient node selection approaches that use global network knowledge are introduced. Performance bounds based on the node/target geometry are derived, and these bounds help to determine the necessary communication reach of the active nodes. The resulting geolocation performance and energy usage, based on communication distance, is evaluated for a decentralized extended Kalman filter (EKF) that is exploiting the different selection approaches. [ABSTRACT FROM AUTHOR]

This paper discusses a new localized resource manager for a wireless sensor network of bearings-only sensors. Specifically, each node uses knowledge of the target under surveillance to determine whether it should actively collect measurements and how far to disseminate the data in order for the sensor network to maintain track of the target. At each node, the resource manager requires only knowledge of the relative location to the target for itself and the active nodes from the previous snapshot. The decentralized strategy represents a modification to the global node selection (GNS) method that exploits knowledge of the location of all nodes in the network. Simulations show that despite the lack of global network knowledge, the new localized management method is almost as effective as GNS in terms of balancing the tradeoff between energy usage and localization accuracy. [ABSTRACT FROM AUTHOR]

We develop a novel pose estimation technique for synthetic aperture radar image chips based on the 2-D continuous wavelet transform (CWT). The pose estimator exploits a fast approximation of the 2-D CWT so that its computational complexity is comparable to a principle component analysis (PCA) based approach. By incorporating energy from various scales, the CWT approach is more robust than the PCA method. Experiments over the MSTAR database confirm the robustness of the CWT method. [Copyright &y& Elsevier]

Introduces a multiscale prescreener algorithm that runs as part of the image formation processing step for ultrawideband (UWB) synthetic aperture radar (SAR) systems. Overview of SAR; Review of standard and quadtree backprojection methods; Description of the scattering model and resulting prescreener algorithm; Performance of the pruned image formation scheme in terms of false alarms.

Focuses on a study which examined the multiplicative speckle models for template-based signal aperture radar (SAR) automatic target recognition (ATR) systems. SAR image models; Validation of multiplicative model; Analysis of different multiplicative models; Conclusions.

Evaluates the utility of the extended fractal feature for the enhancement of the focus of attention stage of a synthetic aperture radar automatic target recognition system. Description of the general framework of target detection algorithms; Definition and properties of extended fractal.

Presents a study which evaluated the effectiveness of multiscale Hurst parameters as features for texture classification and segmentation. Classification accuracy for the generalized Hurst; Evaluation of the segmentation accuracy using generalized and standard Hurst features; Effectiveness of the Hurst features to segment real synthetic aperture radar (SAR) imagery.

As you may know, the M Barry Carlton Award is bestowed on the authors of the best paper published in the IEEE Transactions of Aerospace and Electronic Systems (T-AES) for a given calendar year. The winners for the 2011 Award were announced in the February issue earlier this year. It is now my privilege to announce that the 2012 M Barry Carlton award winners are Kevin (Jim) Sangston, Fulvio Gini, and Maria (Sabrina) Greco for their paper entitled, "Coherent Radar Target Detection in Heavy-Tailed Compound-Gaussian Clutter." [ABSTRACT FROM AUTHOR]

This is the last issue of the year, and the Transactions is making great strides to improve the publication timelines. We are in the process of upgrading the web review system to incorporate processing manuscripts post acceptance. This will provide a convenient portal for Rita Janssen, our industrious adminstrative editor, and authors to remain in communication through the entire process up to publication. Furthermore, the system should enable rapid posting of the manuscripts onto Xplore, which is an enhancement that many students and junior faculty will appreciate. The rapid post will be considered a "publication" even though all the editing and typesetting is not complete. For the researcher, this will mean that credit for your work will be given earlier. And for the editorial board, this will mean that our time from manuscript submission to publication should be reduced by about 6 months. This enhanced web review system will also allow us to perpetually archive all files related to a published paper in one location. We hope that this upgrade goes live by February 2015. [ABSTRACT FROM AUTHOR]

Presents the introductory editorial for this issue of the publication. [ABSTRACT FROM PUBLISHER]

Presents the introductory editorial for this issue of the publication. [ABSTRACT FROM AUTHOR]

When the new year begins, TAES will start employing IEEE Publication Services for the copyedit and typesetting of our accepted manuscripts. The choice was made on the basis of a number of factors, one of the most important being that we have come to realize that a tighter relationship to IEEE publishing operations will help us reduce the time to post our issues on IEEE Xplore and facilitate early posting of manuscripts. I know that many authors anxiously wait to see their articles published. I am sure both authors and readers will appreciate the timely appearance of our issues; and our impact will almost certainly be enhanced. [ABSTRACT FROM AUTHOR]

For those authors who have submitted papers recently, you may have noticed some changes in our submission process. As of July 1, IEEE now requires that all authors provide an ORCID at submission. ORCID is short for Open Researcher and Contributor ID, and it serves as a unique identifying number for a researcher so that the various electronic archives can properly attribute authorship of papers to the correct people. This can be thought of as a government identification number that persists for one?s lifetime (like a social security number in the US). It is being put in place to solve the entity resolution problem. [ABSTRACT FROM PUBLISHER]

It is revealed that there is a link between the quantization approach employed and the dimension of the vector parameter which can be accurately estimated by a quantized estimation system. A critical quantity called inestimable dimension for quantized data (IDQD) is introduced, which does not depend on the quantization regions and the statistical models of the observations but instead depends only on the number of sensors and on the precision of the vector quantizers employed by the system. It is shown that the IDQD describes a quantization-induced fundamental limitation on the estimation capabilities of the system. To be specific, if the dimension of the desired vector parameter is larger than the IDQD of the quantized estimation system, then the Fisher information matrix for estimating the desired vector parameter is singular, and, moreover, there exist infinitely many nonidentifiable vector parameter points in the vector parameter space. Furthermore, it is shown that under some common assumptions on the statistical models of the observations and the quantization system, a smaller IDQD can be obtained, which can specify an even more limiting quantization induced fundamental limitation on the estimation capabilities of the system. [ABSTRACT FROM AUTHOR]

It is my pleasure to announce the winners of the 2011 M Barry Carlton Award. These winners authored what is deemed to be the ?best? paper published in IEEE Transactions on Aerospace and Electronic Systems during calendar year 2011. [ABSTRACT FROM AUTHOR]

Change is inevitable, and this makes life both interesting and challenging. Many changes have occurred within these Transactions since the turn of the century. Like most all other journals, we have implemented an electronic web review system. In the old days, correspondences between authors, editors and reviewers relied on the postal service. The processes of generating these correspondences were tedious because authors had to mail out multiple copies of their papers that the editors then sorted and mailed out to reviewers along with the appropriate review forms. Most times, the editors had no idea whether or not the reviewers would respond. These days, the web review system automates a good chunk of this process. In fact, the EIC can quickly look over the status of every paper in the system and take over if there are any problems. In the old days, the manuscripts were sent directly by the authors to the appropriate technical editor, and the EIC had no reliable way to track the papers. This meant that the EIC had to rely on and trust in the efficacy of the technical editors. If, for some reason, a technical editor stopped performing their TAES duties, for whatever reason, months could go by before the EIC realized there was a problem, and by then, the molehills had turned into towering peaks. Thus, it was essential for the EIC to vet and hold onto every member of his/her editorial staff as much as possible. [ABSTRACT FROM AUTHOR]

This correspondence corrects a minor error in the description of maximum likelihood (ML) estimators that extract angle measurements from two unresolved targets using measurements from a monopulse radar. [ABSTRACT FROM AUTHOR]

In this day and age, we have an incredible wealth of knowledge available at our fingertips. This is also increasing the rate of new discoveries. The amount of content in our knowledge repositories such as in IEEE Xplore is growing exponentially. While the easy access to knowledge is helping to drive this growth, there are other less wondrous forces that are at play. In fact, the amount of knowledge within Xplore might be growing far less than the content, and IEEE is very concerned that many duplicative papers are in the archives. This may have a detrimental effect on the enthusiasm for subscriptions that IEEE can expect from libraries, and it is libraries that provide the critical financial support of IEEE?s scholarly operations. Furthermore, duplicative content means that the precious time of valuable volunteer editors and reviewers is being wasted. [ABSTRACT FROM PUBLISHER]

The IEEE T-AES just had its five year review by the IEEE Publications Review Advisory Committee (PRAC). This is the opportunity for the IEEE PRAC to learn how we are operating, what we are doing well and what we might not be doing as well. The review also provides an opportunity for the PRAC to communicate some best practices that they learned from other IEEE publications; and, also, for other IEEE publications to learn from us. [ABSTRACT FROM AUTHOR]

It is hard to believe that we are halfway through the twenty-tens (or whatever one likes to call this decade). T-AES is continuing to change with the times. Many of the more mature contributors will remember the time when you submitted your paper to T-AES ? or any journal ? and did not expect to hear from the associate editor about the first review decision until at least a year had passed. [ABSTRACT FROM AUTHOR]

We are now at the halfway point through the year. As many of you are aware, our articles are not being posted on IEEE Xplore in a timely manner. [ABSTRACT FROM AUTHOR]

In this article, the author discusses the achievements of editors at the Institute of Electrical & Electronics Engineers (IEEE) Transactions on Aerospace and Electronic Systems. He highlights the contribution of editors Marina Ruggieri and Yu (Jade) Morton in the research for the advancement of aerospace and electronic systems. He further talks about awards and achievements made by both editors in their lifetime.

Estimation of an unknown deterministic vector from quantized sensor data is considered in the presence of spoofing attacks, which alter the data presented to several sensors. Contrary to the previous work, a generalized attack model is employed which manipulates the data using transformations with arbitrary functional forms determined by some attack parameters whose values are unknown to the attacked system. For the first time, necessary and sufficient conditions are provided under which the transformations provide a guaranteed attack performance in terms of Cramer–Rao Bound (CRB) regardless of the processing the estimation system employs, thus defining a highly desirable attack. Interestingly, these conditions imply that, for any such attack when the attacked sensors can be perfectly identified by the estimation system, either the Fisher information matrix (FIM) for jointly estimating the desired and attack parameters is singular or that the attacked system is unable to improve the CRB for the desired vector parameter through this joint estimation even though the joint FIM is nonsingular. It is shown that it is always possible to construct such a highly desirable attack by properly employing a sufficiently large dimension attack vector parameter relative to the number of quantization levels employed, which was not observed previously. To illustrate the theory in a concrete way, we also provide some numerical results which corroborate that under the highly desirable attack, attacked data are not useful in reducing the CRB. [ABSTRACT FROM AUTHOR]

Network traffic data analysis is important for securing our computing environment and data. However, analyzing network traffic data requires tremendous effort because of the complexity of continuously changing network traffic patterns. To assist the user in better understanding and analyzing the network traffic data, an interactive web-based visualization system is designed using multiple coordinated views, supporting a rich set of user interactions. For advancing the capability of analyzing network traffic data, feature extraction is considered along with uncertainty quantification to help the user make precise analyses. The system allows the user to perform a continuous visual analysis by requesting incrementally new subsets of data with updated visual representation. Case studies have been performed to determine the effectiveness of the system. The results from the case studies support that the system is well designed to understand network traffic data by identifying abnormal network traffic patterns. [ABSTRACT FROM AUTHOR]

The proliferation of smart phones with sensing capabilities motivates exploring the applicability limits of (phone-based) mobile sensing. While a phone can directly measure variables such as location, acceleration, and orientation, other interesting quantities one may want to measure have higher-level semantics that a phone does not directly recognize. For example, one might want to map parking lots that are free after hours, or restaurants that are popular after midnight. How can we measure such higher-level logical quantities using sensors on phones? Techniques that address this question fall in the broad area of surrogate sensing, defined as inferring high-level logical quantities by measuring weaker surrogates. The surrogates in question are variables that can be sensed using a phone, but are only weakly related to the original high-level logical quantities one is really after. The key challenge is to exploit appropriate aggregation techniques that leverage the availability of large numbers of phones to overcome the poor quality of individual surrogates. Recently, significant advances have been made in understanding the quality limits of surrogate sensing. This article overviews the main ideas and insights underlying these advances. [ABSTRACT FROM PUBLISHER]

This article proposes unsupervised truth-finding algorithms that combine consideration of multi-modal content features with analysis of propagation patterns to evaluate the veracity of observations in social sensing applications. A key social sensing challenge is to develop effective algorithms for estimating both the reliability of sources and the veracity of their observations without prior knowledge. In contrast to prior solutions that use labeled examples to learn content features that are correlated with veracity, our approach is entirely unsupervised. Hence, given no prior training data, we jointly learn the importance of different content features together with the veracity of observations using propagation patterns as an indicator of perceived content reliability. A novel penalized expectation maximization (PEM) algorithm is proposed to improve the quality of estimation results for observations bolstered by multiple features. In addition, we develop a constrained expectation maximum likelihood with multiple features (CEM-MultiF) that introduces a novel constraint to boost the probability of correctness of some claims. Finally, we evaluate the performance of the proposed algorithms, called EM-Multi, CEM-Multi and PEM-MultiF, respectively, on real-world data sets collected from Twitter. The evaluation results demonstrate that the proposed algorithms outperform the existing fact-finding approaches, and offer tunable knobs for controlling robustness/performance trade-offs in the presence of malicious sources. [ABSTRACT FROM AUTHOR]

An in-depth understanding of uncertainty is the first step to making effective decisions under uncertainty. Machine/deep learning (ML/DL) has been hugely leveraged to solve complex problems involved with processing high-dimensional data. However, reasoning and quantifying different uncertainties to achieve effective decision-making have been much less explored in ML/DL than in other Artificial Intelligence (AI) domains. In particular, belief/evidence theories have been studied in Knowledge representation and reasoning (KRR) since the 1960s to reason and measure uncertainties to enhance decision-making effectiveness. Based on our in-depth literature review, only a few studies have leveraged mature uncertainty research in belief/evidence theories in ML/DL to tackle complex problems under different types of uncertainty. Our present survey paper discusses major belief theories and their core ideas dealing with uncertainty causes and types and quantifying them, along with the discussions of their applicability in ML/DL. Particularly, we discuss three main approaches leveraging belief theories in Deep Neural Networks (DNNs), including Evidential DNNs, Fuzzy DNNs, and Rough DNNs, in terms of their uncertainty causes, types, and quantification methods along with their applicability in diverse problem domains. Through an in-depth understanding of the extensive survey on this topic, we discuss insights, lessons learned, limitations of the current state-of-the-art bridging belief theories and ML/DL, and future research directions. This paper conducts an extensive survey by bridging belief theories and deep learning in reasoning and quantifying uncertainty to help researchers initiate uncertainty and decision-making research. • Identifying the causes and types of uncertainty in various belief models. • Uncertainty reasoning and quantification of belief models for decision-making. • Applying uncertainty quantification of belief models into deep learning. • Comparing uncertainty consideration in belief models and deep learning. • Insights, lessons learned, limitations, and future research directions. [ABSTRACT FROM AUTHOR]

Online social media have democratized the broadcasting of information, encouraging users to view the world through the lens of social networks. The exploitation of this lens, termed social sensing, presents challenges for researchers at the intersection of computer science and the social sciences. [ABSTRACT FROM AUTHOR]

We consider a sensor network focused on target localization, where sensors measure the signal strength emitted from the target. Each measurement is quantized to one bit and sent to the fusion center. A general attack is considered at some sensors that attempts to cause the fusion center to produce an inaccurate estimation of the target location. The attack is a combination of man-in-the-middle, hacking, and spoofing attacks that can effectively change both signals going into and coming out of the sensor nodes in a realistic manner. We show that the essential effect of attacks is to alter the naive estimate of the distance between the target and each attacked sensor, which ignores the existence of attacks, to a different extent, giving rise to a geometric inconsistency among the attacked and unattacked sensors. With the help of two secure sensors, a class of detectors are proposed to detect the attacked sensors by scrutinizing the existence of the geometric inconsistency.We show that the false alarm and miss probabilities of the proposed detectors decrease exponentially as the number of measurement samples increases, which implies that with sufficient measurement samples, the proposed detectors can identify the attacked and unattacked sensors with any required accuracy. Numerical results show that compared to the cases where all sensors are employed without detecting attacks or only the secure sensors are employed, the localization performance can be significantly improved if we employ the secure sensors and the sensors which are declared as unattacked by the proposed detector. [ABSTRACT FROM AUTHOR]

The real-time discovery of local events (e.g., protests, disasters) has been widely recognized as a fundamental socioeconomic task. Recent studies have demonstrated that the geo-tagged tweet stream serves as an unprecedentedly valuable source for local event detection. Nevertheless, how to effectively extract local events from massive geo-tagged tweet streams in real time remains challenging. To bridge the gap, we propose a method for effective and real-time local event detection from geo-tagged tweet streams. Our method, named GeoBurst+, first leverages a novel cross-modal authority measure to identify several pivots in the query window. Such pivots reveal different geo-topical activities and naturally attract similar tweets to form candidate events. GeoBurst+ further summarizes the continuous stream and compares the candidates against the historical summaries to pinpoint truly interesting local events. Better still, as the query window shifts, GeoBurst+ is capable of updating the event list with little time cost, thus achieving continuous monitoring of the stream. We used crowdsourcing to evaluate GeoBurst+ on two million-scale datasets and found it significantly more effective than existing methods while being orders of magnitude faster. [ABSTRACT FROM AUTHOR]

In this paper, we present GDA, a generalized decision aggregation framework that integrates information from distributed sensor nodes for decision making in a resource efficient manner. Different from traditional approaches, our proposed GDA framework is able to not only estimate the reliability of each sensor, but also take advantage of its confidence information, and thus achieves higher decision accuracy. Targeting generalized problem domains, our framework can naturally handle the scenarios where different sensor nodes observe different sets of events whose numbers of possible classes may also be different. GDA also makes no assumption about the availability level of ground truth label information, while being able to take advantage of any if present. For these reasons, our approach can be applied to a much broader spectrum of sensing scenarios. In this paper, we also propose two extensions of the GDA framework, i.e., incremental GDA (I-GDA) and parallel GDA (P-GDA) to deal with streaming and large-scale data. The advantages of our proposed methods are demonstrated through both theoretic analysis and extensive experiments. [ABSTRACT FROM PUBLISHER]

Detecting complex events from subsymbolic data streams (such as images, audio recordings or videos) is a challenging problem, as traditional symbolic approaches cannot be used to process subsymbolic data, and neural-only approaches usually require larger amounts of training data than available. In this paper, we present DeepProbCEP, a Complex Event Processing (CEP) approach designed with four objectives: (i) allowing the use of subsymbolic data as an input, (ii) retaining flexibility and modularity in the definition of complex event rules, (iii) limiting the cost of obtaining training data and (iv) being robust against adversarial conditions. DeepProbCEP archives this by using a neuro-symbolic approach, which combines the neural and symbolic approaches to allow training with sparse data. This is made possible through the injection of human knowledge. In this paper, we demonstrate that DeepProbCEP outperforms other state-of-the-art approaches when training using sparse data. We also show that DeepProbCEP is robust in different adversarial settings. Finally, DeepProbCEP's flexibility is demonstrated by showing it can be used to process both images and audio as input. • A neuro-symbolic approach for Complex Event Processing on subsymbolic data. • Our approach provides flexibility when defining the rules for complex events. • Our approach reduces the cost of obtaining training data. • Our approach is robust against adversarial conditions. [ABSTRACT FROM AUTHOR]

We study a problem in which a single sensor is scheduled to observe sites periodically, motivated by applications in which the goal is to maintain up-to-date readings for all the observed sites. In the existing literature, it is typically assumed that the time for a sensor switching from one site to another is negligible. This may not be the case in applications such as camera surveillance of a border, however, in which the camera takes time to pan and tilt to refocus itself to a new geographical location. We formulate a problem with constraints modeling refocusing delays. We prove the problem to be NP-hard and then study a special case in which refocusing is proportional to some Euclidian metric. We give a lower bound on the optimal cost for the scheduling problem, and we derive exact solutions for some special cases of the problem. Finally, we provide and experimentally evaluate several heuristic algorithms, some of which are based on the computed lower bound, for the setting of one sensor and many sites. [ABSTRACT FROM AUTHOR]