96 results found
Yang P, Kolbeinsson A, Shukla N, et al., 2023, Deep contrastive anomaly detection for airline ancillaries prediction, 2022 21th IEEE International Conference on Machine Learning and Applications (ICMLA), Publisher: IEEE, Pages: 1167-1174
The increasing range of ancillary products offered by airlines is making existing, static frameworks obsolete. The changing expectations of customers have created a need for more dynamic and reactive offers. In order to tailor an offer to an individual journey, it is possible to leverage similar journeys and the observed outcomes in a semi-supervised approach.In this paper, a multi-stage deep learning framework, namely Deep Ancillaries Prediction (DAP), is developed to understand personalised demand for airline ancillaries and improve pricing strategies. DAP aims to solve the overlapping distribution problem and class imbalances observed in real-world airlinedatasets. The framework incorporates a contrastive learning module to learn richer feature embeddings and an autoencoder for semi-supervised learning into one framework, and outperforms current ancillary prediction systems. The modules can be trained separately and hence, are suitable for an online learning setting. This framework is designed to be transferable to differentprediction tasks in the airline industry. Significant performance enhancements are attained compared to the current state-of-the-art algorithms.
Garcia-Trevino E, Yang P, Barria J, 2022, Wavelet probabilistic neural networks, IEEE Transactions on Neural Networks and Learning Systems, ISSN: 1045-9227
In this article, a novel wavelet probabilistic neural network (WPNN), which is a generative-learning wavelet neural network that relies on the wavelet-based estimation of class probability densities, is proposed. In this new neural network approach, the number of basis functions employed is independent of the number of data inputs, and in that sense, it overcomes the well-known drawback of traditional probabilistic neural networks (PNNs). Since the parameters of the proposed network are updated at a low and constant computational cost, it is particularly aimed at data stream classification and anomaly detection in off-line settings and online environments where the length of data is assumed to be unconstrained. Both synthetic and real-world datasets are used to assess the proposed WPNN. Significant performance enhancements are attained compared to state-of-the-art algorithms.
Tu G, Junyent Ferre A, Xiang J, et al., 2021, Optimal power sharing of wind farms for frequency response, IET Renewable Power Generation, Vol: 15, Pages: 1005-1018, ISSN: 1752-1416
This paper presents a uniform optimal power sharing strategy to coordinate the wind turbines (WTs) in a wind farm (WF)to provide occasional and continuous frequency response (FR). The coordination of WTs is formulated as an optimisation problemthat takes into account the WT dynamics and tries to reduce the long term loss of energy yield caused by the provision of FR. Thisis achieved by maximising the total kinetic energy of the WF over time while reducing wear and tear of WTs. The proposed optimalpower sharing strategy relies on periodic communication between each WT and a WF controller. Local linear approximations areemployed to predict the system behaviour and the solution of the optimisation problem is obtained using the proposed centralisedand/or distributed algorithm. The distributed algorithm only requires one-way communication between the WF controller and localWTs, reducing the communication overheads. Simulation studies are carried out on a WF model to demonstrate the effectivenessof the proposed strategy. The results show the strategy enables reduction of yield loss over previous methods while avoiding overtorque operation during FR provision.
Garcia-Trevino E, Alarcon-Aquino V, Barria J, 2019, The radial wavelet frame density estimator, Computational Statistics & Data Analysis, Vol: 130, Pages: 111-139, ISSN: 0167-9473
The estimation of probability densities is one of the fundamental problems in scientific research. It has been shown that Wavelet Density Estimators, which are a well-documented nonparametric approach, outperform other nonparametric estimators in problems involving densities with discontinuities and local features. However, the use of this type of estimators is not widely extended in the scientific community mainly because of their heavy computational complexity and their difficult algorithmic implementation. A novel multidimensional Wavelet Density Estimator approach based on new multidimensional scaling functions with analytic closed-form expressions is proposed. The key advantages of the proposed estimator are its simpler multidimensional algorithmic implementation and its significant reduction in computational complexity. Algorithmic formulations for four different data analysis scenarios are presented: (1) batch processing of input data, (2) online estimation for stationary process, (3) online estimation for non-stationary contexts and (4) batch estimation of high-dimensional data. The assessment results show that the proposed approach reduces the computational time of the estimation process while maintaining competitive estimation errors.
Garcia-Trevino E, Alarcon-Aquino V, Barria J, 2018, The Radial Wavelet Frame Density Estimator
R code for all the proposed algorithms: The Radial Wavelet Frame Density Estimator
Garcia-Trevino E, Hameed MZ, Barria JA, 2018, Data stream evolution diagnosis using recursive wavelet densityestimators, ACM Transactions on Knowledge Discovery from Data, Vol: 12, Pages: 1-28, ISSN: 1556-4681
Data streams are a new class of data that is becoming pervasively important in a wide range of applications, ranging from sensor networks, environmental monitoring to finance. In this article, we propose a novel framework for the online diagnosis of evolution of multidimensional streaming data that incorporates Recursive Wavelet Density Estimators into the context of Velocity Density Estimation. In the proposed framework changes in streaming data are characterized by the use of local and global evolution coefficients. In addition, we propose for the analysis of changes in the correlation structure of the data a recursive implementation of the Pearson correlation coefficient using exponential discounting. Two visualization tools, namely temporal and spatial velocity profiles, are extended in the context of the proposed framework. These are the three main advantages of the proposed method over previous approaches: (1) the memory storage required is minimal and independent of any window size; (2) it has a significantly lower computational complexity; and (3) it makes possible the fast diagnosis of data evolution at all dimensions and at relevant combinations of dimensions with only one pass of the data. With the help of the four examples, we show the framework’s relevance in a change detection context and its potential capability for real world applications.
Milojevic M, Barria J, 2017, Early warnings dissemination for urban micro-scale monitoring using vehicular sensor network, 5th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS 2017), Publisher: IEEE, Pages: 244-249
This paper presents the beaconless multi-hop Decentralised Dissemination of Warnings (DDW) mechanism for micro-scale monitoring applications in urban environments. It is designed for a Vehicular Sensor Network (VSN) based on mobile nodes with limited resources. Mobile nodes sense and generate high pollution levels early warnings in near real-time. The warning messages are disseminated toward Static Monitoring Units used by the local authorities to monitor onsets of harmful pollution level episodes. The DDW is a distance-based mechanism where a receiving node calculates the waiting time before rebroadcasting the message based on the distance from the sending node and the distance to the Static Monitoring Units. The DDW mechanism performance is evaluated in an urban environment with different number of mobile nodes in the network. Results show that the proposed mechanism collects more non-duplicated warnings at the Static Monitoring Units than other evaluated dissemination protocols. It is also shown the DDW mechanism reduces the amount of duplicated warning messages sent in the network, which is especially important when mobile nodes are resource constrained.
milojevic M, barria J, 2017, Decentralized data fusion for urban micro-scale monitoring using mobile sensor network, International Conference on Networked Systems GI-ITG (NetSys'17), Publisher: IEEE
This paper presents a Decentralized Data Fusion (DDF) framework for micro-scale monitoring applications in urban environments using a mobile sensor network. Here nodes collect data along their routes and share them with other nodes in the network in an opportunistic manner. The DDF framework enables the nodes to fuse data that arrives delayed from other nodes and estimate the missing values in the time gaps. This allows the nodes to create an autonomous perception about the dynamics of the observed phenomenon. The performance of the proposed DDF framework is demonstrated in the context of an urban air pollution monitoring scenario. Simulation results show that the proposed framework is able to expand the estimated pollution data set at the expense of a slight decrease in its accuracy. The simulation results also evaluate the impact of population of nodes on the performance of the DDF framework.
Milojevic M, Barria J, 2016, Decentralized Data Dissemination and Harvestingfor Urban Monitoring, IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) 2016, Publisher: IEEE, ISSN: 2166-9589
This paper presents a Decentralized dataDissemination and Harvesting (DDH) mechanism for urbanpollution monitoring using mobile sensor nodes with limitedresources. The proposed DDH mechanism enables participatingnodes to self-decide whether to process the received data ornot, thus, reducing the on-board processing load. Based onthe harvested data, the nodes calculate their level of interestin monitoring the particular street segments. In this way areduction in the number of actively participating nodes isaccomplished. In addition, the mobile nodes process raw sensorreadings using the Delayed State Information Filter (DSIF) tomaintain the past pollution states and perform a decentralizeddata fusion. The proposed DDH mechanism is assessed usingsimulations with varying number of the participating nodes.The results show that the proposed mechanism outperformsexisting solutions in terms of the utilisation of nodes resources,without affecting the amount of volume of gathered data for themonitored street segments.
Hamid QR, Barria JA, 2016, Congestion avoidance for recharging electric vehicles using smoothed particle hydrodynamics, IEEE Transactions on Power Systems, Vol: 31, Pages: 1014-1024, ISSN: 0885-8950
In this paper, a novel approach for recharging electric vehicles (EVs) is proposed based on managing multiple discrete units of electric power flow, named energy demand particles (EDPs). Key similarities between EDPs and fluid particles (FPs) are established that allow the use of a smoothed particle hydrodynamics (SPH) method for scheduling the recharging times of EVs. It is shown, via simulation, that the scheduling procedure not only minimizes the variance of voltage drops in the secondary circuits, but it also can be used to implement a dynamic demand response and frequency control mechanism. The performance of the proposed scheduling procedure is also compared with alternative approaches recently published in the literature.
Thajchayapong S, Barria JA, 2015, Spatial inference of traffic transition using micro-macro traffic variables, IEEE Transactions on Intelligent Transportation Systems, Vol: 16, Pages: 854-864, ISSN: 1524-9050
This paper proposes an online traffic inference algorithm for road segments in which local traffic information cannot be directly observed. Using macro-micro traffic variables as inputs, the algorithm consists of three main operations. First, it uses interarrival time (time headway) statistics from upstream and downstream locations to spatially infer traffic transitions at an unsupervised piece of segment. Second, it estimates lane-level flow and occupancy at the same unsupervised target site. Third, it estimates individual lane-level shockwave propagation times on the segment. Using real-world closed-circuit television data, it is shown that the proposed algorithm outperforms previously proposed methods in the literature.
Garcia-Trevino ED, Barria JA, 2014, Structural generative descriptions for time series classification, IEEE Transactions on Cybernetics, Vol: 44, Pages: 1978-1991, ISSN: 1083-4419
In this paper, we formulate a novel time series representation framework that captures the inherent data dependency of time series and that can be easily incorporated into existing statistical classification algorithms. The impact of the proposed data representation stage in the solution to the generic underlying problem of time series classification is investigated. The proposed framework, which we call structural generative descriptions moves the structural time series representation to the probability domain, and hence is able to combine statistical and structural pattern recognition paradigms in a novel fashion. Two algorithm instantiations based on the proposed framework are developed. The algorithms are tested and compared using different publicly available real-world benchmark data. Results reported in this paper show the potential of the proposed representation framework, which in the experiments investigated, performs better or comparable to state-of-the-art time series description techniques.
Kim J, Barria JA, Chang M, et al., 2013, Special Issue on Embedded Systems for Interactive Multimedia Services (ES-IMS), ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS, Vol: 12, ISSN: 1539-9087
Hamid QR, Barria JA, 2013, Distributed Recharging Rate Control for Energy Demand Management of Electric Vehicles, IEEE Transactions on Power Systems, Vol: 28, Pages: 2688-2699, ISSN: 0885-8950
Thajchayapong S, Garcia-Trevino ES, Barria JA, 2013, Distributed Classification of Traffic Anomaliesusing Microscopic Traffic Variables, IEEE Transactions on Intelligent Transportation Systems, Vol: 1, Pages: 448-458, ISSN: 1524-9050
Yang Q, Laurenson DI, Barria JA, 2012, On the use of LEO satellite constellation for active network management in power distribution networks, IEEE Transactions on Smart Grid, Vol: 3, Pages: 1371-1381, ISSN: 1949-3053
The passive nature of power distribution networkshas been changing to an active one in recent years as the number of small-scale Distributed Generators (DGs) connected to them rises. The consensus of recent research is that currentslow central network control based upon Supervisory Control and Data Acquisition (SCADA) systems is no longer sufficient and Distributed Network Operators (DNOs) wish to adopt novelmanagement mechanisms coupled with advanced communication infrastructures to meet the emerging control challenges. In thispaper, we address this issue from the communication perspective by exploiting the effectiveness of using a Low Earth Orbit (LEO)satellite network as the key component of the underlying communication infrastructure to support a recently suggested activenetwork management solution. The key factors that would affect the communication performance over satellite links are discussed and an analytical LEO network model is presented. The delivery performance of several major data services for supporting the management solution is evaluated against a wide range of satellitelink delay and loss conditions under both normal and emergency traffic scenarios through extensive simulation experiments. Our investigation demonstrates encouraging results which suggests that a LEO network can be a viable communication solution for managing the next-generation power energy networks.
Lent R, Minero M, North R, et al., 2012, Evaluating Mobility Models in Participatory Sensing, ACM international workshop on Mission-oriented wireless sensor networking, Publisher: ACM, Pages: 3-8
Yang Q, Barria JA, Green TC, 2011, Communication Infrastructures for DistributedControl of Power Distribution Networks, IEEE Transactions on Industrial Informatics, Vol: 7, Pages: 316-327, ISSN: 1551-3203
Lent R, Barria J, 2011, Towards Reliable Mobile Ad Hoc Networks, Mobile Ad-hoc Networks: Protocol Design, Pages: 99-120, ISBN: 978-953-307-402-3
Garcia-Trevino E, Barria JA, 2011, Online wavelet-based density estimation for non stationary streaming data, Computational Statistics and Data Analysis, Vol: 56, Pages: 327-344, ISSN: 0167-9473
Barria JA, Thajchayapong S, 2011, Detection and classification of traffic anomalies using microscopic traffic variables, IEEE Transactions on Intelligent Transportation Systems, Pages: 1-10, ISSN: 1524-9050
Ahmad M, Alexandrou I, Al-Nuaimy W, et al., 2010, WCE 2010 - World Congress on Engineering 2010: Preface, WCE 2010 - World Congress on Engineering 2010, Vol: 2
Thajchayapong S, Barria JA, Garcia-Trevino E, 2010, Lane-level traffic estimations using microscopic traffic variables, Proceedings of 13th IEEE Conference on Intelligent Transportation Systems, Publisher: IEEE, Pages: 1189-1194
Thajchayapong S, Barria JA, 2010, Anomaly Detection using Microscopic TrafficVariables on Freeway Segments, Transportation Research Board 89th Annual Meeting
This paper proposes and assesses the effectiveness of monitoring vehicular traffic anomalies usingmicroscopic traffic variables, namely relative speed and inter-vehicle spacing. We present analgorithm that detects transient changes in traffic patterns using microscopic traffic variables. Inparticular, we show that when applied to real-world scenarios, our algorithm can use the varianceof statistics of relative speed to detect traffic anomalies and precursors to non-recurring traffic congestion.The performance of the proposed algorithm is also assessed using a microscopic trafficsimulation environment, where we show that with minimum prior knowledge, the proposed algorithmhas comparable performance to an ideally placed loop detector monitoring the standarddeviation of speed. The algorithm also performs very well even when the microscopic traffic variablesare available only from a fraction of the complete population of vehicles.
Han J, Krishnan R, Polak JW, et al., 2010, A new method for probabilistic traffic state identification using loop detector data: theory and empirical results, 42nd Annual Meeting of the Universities Transport Study Group
Han J, Polak JW, Barria J, et al., 2010, On the estimation of space-mean-speed from inductive loop detector data, TRANSPORTATION PLANNING AND TECHNOLOGY, Vol: 33, Pages: 91-104, ISSN: 0308-1060
Regner T, Barria J, Pitt J, et al., 2010, Governance of Digital Content in the Era of Mass Participation, Electronic Commerce Research, Vol: 10, Pages: 99-110, ISSN: 1389-5753
Regner T, Barria J, Pitt J, et al., 2010, Governance of digital content in the era of mass participation, Electronic Commerce Research, Vol: 10, Pages: 99-110, ISSN: 1389-5753
Regner T, Barria JA, Pitt JV, et al., 2009, An artist life cycle model for digital media content: Strategies for the Light Web and the Dark Web, ELECTRONIC COMMERCE RESEARCH AND APPLICATIONS, Vol: 8, Pages: 334-342, ISSN: 1567-4223
Lent R, Barria JA, 2009, Sensor-Aided Routing for Mobile Ad Hoc Networks, Wireless Sensor Networks Symposium (IWCMC)
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.