Search

Existing macroeconomic forecasting methods primarily focus on the characteristics of economic data, but they overlook the energy-related features concealed behind these economic characteristics, which may lead to inaccurate GDP predictions. Therefore, this paper meticulously analyzes the relationship between energy big data and economic data indicators, explores the coupling feature mining of energy big data and economic data, and constructs features coupling economic and energy data. Targeting the nonlinear variation coupling features in China's quarterly GDP data and using the long short-term memory (LSTM) neural network model based on deep learning, we employ wavelet analysis technology (WA) to decompose selected macroeconomic variables and construct a prediction model combining LSTM and WA, which is further compared with multiple benchmark models. The research findings show that, in terms of quarterly GDP data prediction, the combined deep learning model and wavelet analysis significantly outperform other methods. When processing structurally complex, nonlinear, and multi-variable data, the LSTM and WA combined prediction model demonstrate better generalization capabilities, with its prediction accuracy generally surpassing other benchmark models. [ABSTRACT FROM AUTHOR]

This paper is an attempt to bridge the gap between deep learning and grammatical inference. Indeed, it provides an algorithm to extract a (stochastic) formal language from any recurrent neural network trained for language modelling. In detail, the algorithm uses the already trained network as an oracle—and thus does not require the access to the inner representation of the black-box—and applies a spectral approach to infer a weighted automaton. As weighted automata compute linear functions, they are computationally more efficient than neural networks and thus the nature of the approach is the one of knowledge distillation. We detail experiments on 62 data sets (both synthetic and from real-world applications) that allow an in-depth study of the abilities of the proposed algorithm. The results show the WA we extract are good approximations of the RNN, validating the approach. Moreover, we show how the process provides interesting insights toward the behavior of RNN learned on data, enlarging the scope of this work to the one of explainability of deep learning models. [ABSTRACT FROM AUTHOR]

The analysis of the vascular tree is a fundamental part of the clinical assessment of retinal images. The diversity of blood vessel calibers and curvatures, as well as the ocular vascular alterations derived from the progress of certain diseases, make automatic blood vessel segmentation a challenging task. In this paper, a novel Width Attention-based Convolutional Neural Network, called WA-Net, is proposed. In the WA-Net, a fundus image is decomposed into multiple channels by a layer of Distorted Second-Order Differential Gaussian Matched Filters (DSD-GMF), where each channel is associated with a blood-vessel width. Subsequently, the channel relevance is weighted through the Width Attention Module (WAM), which considers channel and position correlations. Finally, in order to specialize the feature maps with a concrete vessel-width category, either thin-vessel or thick-vessel related, the weighted channels are divided into two groups by the Two-Stream Block, composed of three-level UNet streams. Experimental results on three public datasets (DRIVE/STARE/CHASE) indicate that the proposed method provides a performance gain over other attention-based and non-attention-based architectures, achieving state-of-the-art Accuracy and Area-Under-the-Receiver-Operating-Characteristic-Curve scores (0.9575/0.9665/9653 and 0.9784/0.9865/0.9841 within the Field-of-View, respectively). [Display omitted] • A Width-Aware model improves the state-of-the-art for Retinal Vessel Segmentation. • WA-Net handles blood vessel width variety through a Gaussian-Matched-Filter layer. • WA-Net specializes feature extraction through using a Two-Stream scheme. • WA-Net is more accurate and 13x smaller than other state-of-the-art models. [ABSTRACT FROM AUTHOR]

The solar ultraviolet index (UVI) is a key public health indicator to mitigate the ultraviolet-exposure related diseases. This study aimed to develop and compare the performances of different hybridised deep learning approaches with a convolutional neural network and long short-term memory referred to as CLSTM to forecast the daily UVI of Perth station, Western Australia. A complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is incorporated coupled with four feature selection algorithms (i.e., genetic algorithm (GA), ant colony optimization (ACO), particle swarm optimization (PSO), and differential evolution (DEV)) to understand the diverse combinations of the predictor variables acquired from three distinct datasets (i.e., satellite data, ground-based SILO data, and synoptic mode climate indices). The CEEMDAN-CLSTM model coupled with GA appeared to be an accurate forecasting system in capturing the UVI. Compared to the counterpart benchmark models, the results demonstrated the excellent forecasting capability (i.e., low error and high efficiency) of the recommended hybrid CEEMDAN-CLSTM model in apprehending the complex and non-linear relationships between predictor variables and the daily UVI. The study inference can considerably enhance real-time exposure advice for the public and help mitigate the potential for solar UV-exposure-related diseases such as melanoma. [ABSTRACT FROM AUTHOR]

This work determines whether hyperspectral imaging is suitable for discriminating ore from waste at the point of excavation. A prototype scanning system was developed for this study. This system combined hyperspectral cameras and a three-dimensional LiDAR, mounted on a pan-tilt head, and a positioning system which determined the spatial location of the resultant hyperspectral data cube. This system was used to obtain scans both in the laboratory and at a gold mine in Western Australia. Samples from this mine site were assayed to determine their gold concentration and were scanned using the hyperspectral apparatus in the laboratory to create a library of labelled reference spectra. This library was used as (i) the reference set for spectral angle mapper classification and (ii) a training set for a convolutional neural network classifier. Both classification approaches were found to classify ore and waste on the scanned face with good accuracy when compared to the mine geological model. Greater resolution on the classification of ore grade quality was compromised by the quality and quantity of training data. The work provides evidence that an excavator-mounted hyperspectral system could be used to guide a human or autonomous excavator operator to selectively dig ore and minimise dilution. [ABSTRACT FROM AUTHOR]

Deep-learning (DL) postprocessing methods are examined to obtain reliable and accurate probabilistic forecasts from single-member numerical weather predictions of integrated vapor transport (IVT). Using a 34-yr reforecast, based on the Center for Western Weather and Water Extremes West-WRF mesoscale model of North American West Coast IVT, the dynamically/statistically derived 0–120-h probabilistic forecasts for IVT under atmospheric river (AR) conditions are tested. These predictions are compared with the Global Ensemble Forecast System (GEFS) dynamic model and the GEFS calibrated with a neural network. In addition, the DL methods are tested against an established, but more rigid, statistical–dynamical ensemble method (the analog ensemble). The findings show, using continuous ranked probability skill score and Brier skill score as verification metrics, that the DL methods compete with or outperform the calibrated GEFS system at lead times from 0 to 48 h and again from 72 to 120 h for AR vapor transport events. In addition, the DL methods generate reliable and skillful probabilistic forecasts. The implications of varying the length of the training dataset are examined, and the results show that the DL methods learn relatively quickly and ∼10 years of hindcast data are required to compete with the GEFS ensemble. [ABSTRACT FROM AUTHOR]

Quantification of halloysite and kaolinite in clay deposits from X-ray diffraction (XRD) commonly requires extensive sample preparation to differentiate the two phyllosilicates. When assessing hundreds of samples for mineral resource estimations, XRD analyses may become unfeasible due to time and expense. Fourier transform infrared (FTIR) analysis is a fast and cost-effective method to discriminate between kaolinite and halloysite; however, few efforts have been made to use this technique for quantified analysis of these minerals. In this study, we trained machine- and deep-learning models on XRD data to predict the abundance of kaolinite and halloysite from FTIR, chemical composition, and brightness data. The case study is from the Cloud Nine kaolinite–halloysite deposit, Noombenberry Project, Western Australia. The residual clay deposit is hosted in the saprolitic and transition zone of the weathering profile above the basement granite on the southwestern portion of the Archean Yilgarn Craton. Compared with XRD quantification, the predicted models have an R2 of 0.97 for kaolinite and 0.96 for halloysite, demonstrating an excellent fit. Based on these results, we demonstrate that our methodology provides a cost-effective alternative to XRD to quantify kaolinite and halloysite abundances. [ABSTRACT FROM AUTHOR]