Stock prediction using deep learning

Stock market is considered chaotic, complex, volatile and dynamic. Undoubtedly, its prediction is one of the most challenging tasks in time series forecasting. Moreover existing Artificial Neural Network (ANN) approaches fail to provide encouraging results. Meanwhile advances in machine learning have presented favourable results for speech recognition, image classification and language processing. Methods applied in digital signal processing can be applied to stock data as both are time series. Similarly, learning outcome of this paper can be applied to speech time series data. Deep learning for stock prediction has been introduced in this paper and its performance is evaluated on Google stock price multimedia data (chart) from NASDAQ. The objective of this paper is to demonstrate that deep learning can improve stock market forecasting accuracy. For this, (2D)²PCA + Deep Neural Network (DNN) method is compared with state of the art method 2-Directional 2-Dimensional Principal Component Analysis (2D)²PCA + Radial Basis Function Neural Network (RBFNN). It is found that the proposed method is performing better than the existing method RBFNN with an improved accuracy of 4.8% for Hit Rate with a window size of 20. Also the results of the proposed model are compared with the Recurrent Neural Network (RNN) and it is found that the accuracy for Hit Rate is improved by 15.6%. The correlation coefficient between the actual and predicted return for DNN is 17.1% more than RBFNN and it is 43.4% better than RNN.     

Structural characterization of carboxyl cellulose nanofibers extracted from underutilized sources

Two different chemical methods, TEMPO-oxidation and nitro-oxidation, were used to extract carboxyl cellulose nanofibers(CNFs) from non-wood biomass sources(i.e., jute, soft and hard spinifex grasses). The combined TEMPO-oxidation and homogenization approach was very efficient to produce CNFs from the cellulose component of biomass; however, the nitrooxidation method was also found to be effective to extract CNFs directly from raw biomass even without mechanical treatment.The effect of these two methods on the resulting cross-section dimensions of CNFs was investigated by solution small-angle Xray scattering(SAXS), transmission electron microscopy(TEM) and atomic force microscopy(AFM). The UV-Vis spectroscopic data from 0.1 wt% TEMPO-oxidized nanofiber(TOCNF) and nitro-oxidized nanofiber(NOCNF) suspensions showed that TOCNF had the highest transparency( 95%) because of better dispersion, resulted from the highest carboxylate content(1.2 mmol/g). The consistent scattering and microscopic results indicated that TOCNFs from jute and spinifex grasses possessed rectangular cross-sections, while NOCNFs exhibited near square cross-sections. This study revealed that different oxidation methods can result in different degrees of biomass exfoliation and different CNF morphology.     

Swastika shaped slot embedded two port dual frequency band MIMO antenna for wireless applications

Analog Integrated Circuits and Signal Processing - A compact dual band 2?×?2 multiple-input-multiple-output (MIMO) antenna with dimension...     

RETRACTED ARTICLE: A novel large-scale multimedia image data classification algorithm based on mapping assisted deep neural network

Multimedia Tools and Applications - With the increasing number of the images, how to effectively manage and use these images becomes an urgent problem to be solved. The classification of the images...     

Studies on suitability of wheat flour blends with sweet potato,colocasia and water chestnut flours for noodle making

Blends of refined wheat flour (RWF) with colocasia (CF), sweet potato (SPF) and water chestnut (WCF) flours respectively at replacement level of 25 g/100 g were assessed for their suitability for noodles making. All the native flours as well as their blends with RWF exhibited restricted swelling behavior. Incorporation of SPF or CF into RWF decreased the peak viscosity of flour blends. Noodles prepared from respective flour blends of SPF and CF with RWF showed lower cooking time, higher cooked weight, higher water uptake and higher gruel solid loss in comparison to control sample (RWF noodles). Among blend flours noodles, RWF and SPF blend noodles were rated superior for their organoleptic/eating characteristics of slipperiness, firmness, appearance but undesirable high value of tooth packing. Noodle with acceptable quality characteristics and decreased level of gluten, which may prove beneficial for celiac persons, can be developed using RWF blends with non-conventional flours like colocasia, sweet potato and water chestnut.     

Electrophoretic Characterization and Functional Properties of Rice Proteins from Indian Rice Cultivars

Rice protein isolates and extracts of protein fractions were prepared from Indian rice cultivars, namely, Jaya, HKR-120, P-44, Sharbati, Bas-370, and HBC-19. The protein extracts were characterized using SDS-PAGE. The total protein contents of rice cultivars ranged from 5.46 to 7.02 g/100 g sample with albumin and glutelin fractions showing the highest variability among rice cultivars. The electrophoretic patterns of protein fractions exhibited many varietal differences with glutelin fraction revealing the most heterogeneous (10–17 polypeptide units) and prolamin fraction revealing the most homogenous polypeptide composition (3 polypeptide units). The alkali extracted rice endosperm protein isolates showed favorable emulsifying and foaming capacities particularly at an alkaline pH of 11. The total protein content was significantly correlated positive with foaming capacity (r = 0.917, p < 0.01) and negative with oil absorption capacity (r =??0.914, p < 0.05). The total protein content was also correlated significantly positive with cooking time (r = 0.956, p < 0.01) and cooked grain hardness (r = 0.966, p < 0.01).     

A novel macroeconomic forecasting model based on revised multimedia assisted BP neural network model and ant Colony algorithm

Multimedia Tools and Applications - In this paper, we propose a novel macroeconomic forecasting model based on the revised multimedia assisted BP neural network model and the ant colony algorithm....