The application of deep learning approaches to finance has received a great deal of attention from both investors and researchers. This study presents a novel deep learning framework where wavelet transforms (WT), stacked autoencoders (SAEs) and long-short term memory (LSTM) are combined for stock price forecasting. The SAEs for hierarchically extracted deep features is introduced into stock price forecasting for the first time. The deep learning framework comprises three stages. First, the stock price time series is decomposed by WT to eliminate noise. Second, SAEs is applied to generate deep high-level features for predicting the stock price. Third, high-level denoising features are fed into LSTM to forecast the next day’s closing price. Six market indices and their corresponding index futures are chosen to examine the performance of the proposed model. Results show that the proposed model outperforms other similar models in both predictive accuracy and profitability performance.

In this letter, a novel deep learning framework for hyperspectral image classification using both spectral and spatial features is presented. The framework is a hybrid of principal component analysis, deep convolutional neural networks (DCNNs) and logistic regression (LR). The DCNNs for hierarchically extract deep features is introduced into hyperspectral image classification for the first time. The proposed technique consists of two steps. First, feature map generation algorithm is presented to generate the spectral and spatial feature maps. Second, the DCNNs-LR classifier is trained to get useful high-level features and to fine-tune the whole model. Comparative experiments conducted over widely used hyperspectral data indicate that DCNNs-LR classifier built in this proposed deep learning framework provides better classification accuracy than previous hyperspectral classification methods.

No abstract

Republication or reproduction of this report or its storage and/or dissemination by electronic means isAbstract: Mercury is released into the hydrosphere, atmosphere and biosphere as a consequence of natural and anthropogenic processes. It is cycled in the environment and undergoes transformations of its chemical forms. Although the number of chemical species encountered in environmental and biological samples is not large, the range of matrices and the toxicological significance of mercury, with a consequent need for ultra-trace determination, have resulted in a considerable analytical literature. Sensitive and selective methods for mercury determination, micro methods for the determination of mercury species, the occurrence of mercury species in the various environmental compartments, and analytical considerations, including the availability of certified reference materials, are reviewed.

No abstract

[1] The effect of increased aerosol concentrations on the low-level, non-precipitating, ice-free stratus clouds is examined using a suite of surface-based remote sensing systems. Cloud droplet effective radius and liquid water path are retrieved using cloud radar and microwave radiometer. Collocated measurements of aerosol scattering coefficient, size distribution and cloud condensation nuclei (CCN) concentrations were used to examine the response of cloud droplet size and optical thickness to increased CCN proxies. During the episodic events of increase in aerosol accumulation-mode volume distribution, the decrease in droplet size and increase in cloud optical thickness is observed. The indirect effect estimates are made for both droplet effective radius and cloud optical thickness for different liquid water path ranges and they range 0.02 -0.18 and 0.005 -0.154, respectively. Data are also categorized into thin and thick clouds based on cloud geometric thickness (Dz) and estimates show IE values are relatively higher for thicker clouds.

No abstract

No abstract